Evidence Synthesis______Number 47
Screening for Lipid Disorders in Children and Adolescents: Systematic Evidence Review for the U.S. Preventive Services Task Force
Prepared for: Agency for Healthcare Research and Quality U.S. Department of Health and Human Services 540 Gaither Road Rockville, MD 20850 www.ahrq.gov Contract Number 290-02-0024, Task Order Number 1
Prepared by: Oregon Evidence-based Practice Center Oregon Health and Science University 3181 SW Sam Jackson Park Road Portland, Oregon 97239 www.ohsu.edu/epc/usptf/index.htm
Investigators: Elizabeth M. Haney, MD Laurie Hoyt Huffman, MS Christina Bougatsos, BS Michele Freeman, MPH Rongwei Fu, PhD Robert D. Steiner, MD Mark Helfand, MD, MPH Heidi D. Nelson, MD, MPH
AHRQ Publication No. 07-0598-EF-1 July 2007 DISCLAIMERS
This report is based on research conducted by the Oregon Evidence-based Practice Center (EPC) under contract to the Agency for Healthcare Research and Quality (AHRQ), Rockville, MD (Contract No. 290-02-0024). The findings and conclusions in this document are those of the authors, who are responsible for its content, and do not necessarily represent the views of AHRQ. No statement in this report should be construed as an official position of AHRQ or of the U.S. Department of Health and Human Services.
The information in this report is intended to help clinicians, employers, policymakers, and others make informed decisions about the provision of health care services. This report is intended as a reference and not as a substitute for clinical judgment.
This report may be used, in whole or in part, as the basis for the development of clinical practice guidelines and other quality enhancement tools, or as a basis for reimbursement and coverage policies. AHRQ or U.S. Department of Health and Human Services endorsement of such derivative products may not be stated or implied.
ACKNOWLEDGMENTS
This systematic evidence review was funded by the Agency for Healthcare Research and Quality (AHRQ) for the U.S. Preventive Services Task Force (USPSTF), and the investigators acknowledge the contributions of Janelle Guirguis-Blake, MD, Task Order Officer and Medical Officer, AHRQ. Members of the USPSTF who served as leads for this project include Leon Gordis, MD, DrPH, Carol Loveland-Cherry, PhD, RN, FAAN, Albert Siu, MD, MSPH, and Barbara Yawn, MD, MSc, FAAFP. Investigators thank the following expert reviewers of this report for commenting on draft versions, Peter Belamarich, MD, Shifan Dai, MD, PhD, Steve Daniels, MD, PhD, Roger Illingworth, MD, Patrick McBride, MD, MPH, and Ellis Neufeld, MD, PhD. Investigators also thank Andrew Hamilton, MLS, MS for conducting the literature searches and Miranda Walker, BA for her extensive assistance with the report.
ii
STRUCTURED ABSTRACT
Context. Dyslipidemias, disorders of lipid metabolism, are important risk factors for coronary heart disease (CHD). Identification of children with dyslipidemias could lead to interventions aimed at decreasing their risk of CHD as adults.
Objective. To determine the strengths and limits of evidence about the effectiveness of selecting, testing, and managing children and adolescents with dyslipidemia in the course of routine primary care. Screening children and adolescents has the potential to identify three groups with dyslipidemia: those with 1) undiagnosed monogenic dyslipidemia, 2) undiagnosed secondary dyslipidemia, and 3) idiopathic dyslipidemia (polygenic, risk factor associated, or multifactorial). Key questions examined a chain of evidence about the accuracy and feasibility of screening children in various settings, tracking of lipid levels through childhood to adulthood, role of risk factors in selecting children for screening, effectiveness of interventions for children identified with dyslipidemia, and adverse effects of screening and interventions.
Data Sources. Relevant studies were identified from multiple searches of MEDLINE, PsychInfo, the Cochrane database of systematic reviews and controlled clinical trials, and EMBASE (1966 to September 2005). Additional articles were obtained from recent systematic reviews, reference lists of pertinent studies, reviews, editorials, and websites, and by consulting experts.
Study Selection. Eligible studies were applicable to U.S. clinical practice, available in English, and provided primary data relevant to key questions. Cohort studies were used for assessment of risk factors. Comparative and non-comparative prospective studies of screening for dyslipidemia in children provided information on the efficacy of these programs and the accuracy of screening with family history information. Randomized, non-randomized and non-comparative studies were used for assessment of risk factors. Only randomized controlled trials were considered for examining the effectiveness of interventions. Studies of children with previously diagnosed monogenic dyslipidemia were considered for the evaluation of treatment because those are the only children in whom some drugs have been tested.
Data Extraction. Data were extracted from each study and entered into evidence tables.
Data Synthesis. Studies were summarized by descriptive methods and rated for quality using criteria developed by the U.S. Preventive Services Task Force (USPSTF). Normal values for lipids for children and adolescents are currently defined according to population levels (percentiles). More recent studies indicate age, sex, and racial differences and temporal trends that shift cut points. Tracking of lipid levels through childhood is strongest for TC and LDL. Approximately 40-55% of children with elevated total cholesterol (TC) and low-density lipoprotein (LDL) defined by percentile will continue to have elevated lipids on follow-up. Current screening recommendations based on family history will fail to detect substantial numbers (30-60%) of children with elevated lipids. Evidence from epidemiologic studies establish a strong statistical association between overweight and elevations in lipids whereas other risk factors (diet, physical inactivity, aerobic capacity/fitness, puberty level and smoking) have not been adequately assessed. Currently recommended screening strategies have limited diagnostic accuracy, low adherence to guidelines by providers, and limited compliance by parents and children. No trials compare strategies of screening in children. Parental non- compliance with screening and follow-up recommendations is reported.
Drug treatment for dyslipidemia in children has been studied only in children with familial monogenic dyslipidemias (familial hypercholesterolemia [FH] or familial combined hyperlipidemia [FCH]). In this population, 9 randomized controlled trials demonstrate the effectiveness of statins for reducing TC and LDL (% mean reduction from meta-analysis of trials: 24.4%[95% CI 19.5, 29.2] for TC, 30.8% [95% CI 24.1, 37.5] for LDL, 8 studies). Two fair quality trials showed benefit from bile acid binding resins. Randomized controlled trials of diet supplements (psyllium, oat, garlic extract, and sterol margarine) and advice showed marginal improvements in lipids in children with monogenic dyslipidemia. For children without monogenic dyslipidemia, a good quality study showed that high intensity counseling is effective in reducing TC and LDL levels while the intervention is sustained, but not after it ceases. Other studies of diet advice showed no or minimal improvement. Dietary fiber supplements had mixed results in two trials in children and adolescents without monogenic dyslipidemia, and one oat bran supplement trial showed no effect. Six trials of exercise demonstrated little or no improvements in lipids for children without monogenic dyslipidemia (% mean reduction from meta-analysis of trials: 0% [-5.6, 5.6] for TC, 3.1% [-7.7, 1.5] for LDL, 4 studies).
Eighty-one controlled and non-controlled studies of treatment reported a variety of adverse effects of drug, diet, exercise, and combination therapy in children and adolescents. There are reports of growth retardation and nutritional dwarfing in children and adolescents for whom formal dietary assessment and advice was delayed. Although reported adverse effects were not serious, studies were generally small and not of sufficient duration to determine long-term effects of either short or extended use.
Conclusions. Normal values for lipids for children and adolescents are currently defined according to population levels (percentiles). Tracking of lipid levels in children is variable, although evidence is stronger for TC and LDL than for HDL and TG. Screening using family history misses substantial numbers of children with elevated lipids. Most trials of drug interventions demonstrate improvement, but these trials were performed in selected groups of children. Several key questions could not be addressed because of lack of studies, including the effectiveness of screening on adult CHD or lipid outcomes, optimal ages and intervals for screening children, cost-effectiveness of screening, or the effects of treatment of lipids in childhood on adult CHD outcomes.
Keywords: Dyslipidemia; Children; Screening.
TABLE OF CONTENTS
I. Introduction...... 1 Burden of Condition/Epidemiology...... 2 Familial Hypercholesterolemia...... 3 Familial Combined Hyperlipidemia ...... 3 Familial Defective Apoprotein B (Apo B)...... 3 Familial Hypertriglyceridemia...... 3 Healthcare Interventions ...... 4 Prior Recommendations...... 5 Scope of Evidence Synthesis ...... 6
II. Methods...... 8 Literature Search and Strategy...... 8 Inclusion/Exclusion Criteria ...... 8 Data Extraction and Synthesis ...... 8 Size of Literature Reviewed...... 9 External Review Process...... 9
III. Results ...... 10 Key Question 1. Is screening for dyslipidemia in children effective in delaying the onset and reducing the incidence of CHD-related events? ...... 10 Key Question 2. What is the accuracy of screening for dyslipidemia in identifying children at increased risk of CHD-related events and other outcomes? ...... 10 Key Question 2a. What are abnormal lipid values in children? ...... 10 Summary...... 10 Evidence ...... 10 Key Question 2b. What are the appropriate tests? How well do screening tests (non-fasting total cholesterol, fasting total cholesterol, fasting lipid panel) identify individuals with dyslipidemia? ...... 11 Summary...... 12 Evidence ...... 12 Key Question 2c. How well do lipid levels track from childhood to adulthood?...... 14 Summary...... 14 Evidence ...... 15 Key Question 2d. What is the accuracy of family history in determining risk?...... 16 Summary...... 16 Evidence ...... 16 Key Question 2e. What are other important risk factors?...... 17 Summary...... 17 Evidence ...... 17 Key Question 2f. What are effective screening strategies for children (including frequency of testing, optimal age for testing)?...... 18 Summary...... 18 Evidence ...... 19
v
Neonatal Screening...... 19 Key Question 3. What are the adverse effects of screening including false positives, false negatives, labeling, etc? ...... 20 Summary...... 20 Evidence ...... 20 Key Question 4. In children and adolescents, what is the effectiveness of drug, diet, exercise, and combination therapy in reducing the incidence of adult dyslipidemia, and delaying the onset and reducing the incidence of CHD-related events and other outcomes (including optimal age for initiation of treatment)?...... 21 Key Questions 5 - 8. What is the effectiveness of drug, diet, exercise, and combination therapy for treating dyslipidemia in children/adolescents (including the incremental benefit of treating dyslipidemia in childhood)? ...... 21 Summary...... 21 Evidence ...... 22 Studies in Children with Probable or Definite Familial Hypercholesterolemia ...... 22 Drug Treatment...... 22 Diet Treatment...... 23 Exercise Treatment ...... 24 Studies in Children with Elevated Lipids but Not Meeting Criteria for Familial Hyper- cholesterolemia...... 24 Drug Treatment...... 24 Diet Treatment...... 24 Exercise Treatment ...... 25 Combination Diet and Exercise Treatment...... 25 Key Question 9. What are the adverse effects of drug, diet, exercise, and combination therapy in children/adolescents?...... 26 Summary...... 26 Exercise...... 26 Drug Treatment...... 26 Low-Fat Diet...... 28 Dietary Supplements...... 29 Exercise...... 29 Key Question 10. Does improving dyslipidemia in childhood reduce the risk of dyslipidemia in adulthood? ...... 29 Key Question 11. What are the cost issues involved in screening for dyslipidemia in asymptomatic children?...... 30
IV. Discussion ...... 31 Conclusions...... 31 Limitations of the Literature ...... 33 Future Research ...... 34
References...... 35
vi
Figures Figure 1. Defining the Screening Population Figure 2. National Cholesterol Education Program Guidelines for Risk Assessment and Classification of Children Figure 3. Analytic Framework and Key Questions Figure 4. Probability Distribution for Heterozygous Familial Hypercholesterolemia, Relatives, and the General Population Figure 5a. Studies of Tracking – Total Cholesterol Levels Based on Age at Start of Study Figure 5b. Studies of Tracking – LDL-C Levels Based on Age at Start of Study Figure 5c. Studies of Tracking – HDL-C Levels Based on Age at Start of Study Figure 5d. Studies of Tracking – Triglyceride Levels Based on Age at Start of Study Figure 6a. Studies of Tracking – Total Cholesterol Levels Based on Duration of Study Figure 6b. Studies of Tracking – LDL-C Levels Based on Duration of Study Figure 6c. Studies of Tracking – HDL-C Levels Based on Duration of Study Figure 6d. Studies of Tracking – Triglyceride Levels Based on Duration of Study Figure 7. Meta-analysis of Statin Trials in Children with Familial Hypercholesterolemia: Total Cholesterol Reduction Figure 8. Meta-analysis of Statin Trials in Children with Familial Hypercholesterolemia: LDL-C Reduction Figure 9. Meta-analysis of Statin Trials in Children with Familial Hypercholesterolemia: HDL-C Increase Figure 10. Meta-analysis of Exercise Trials in Children: Total Cholesterol Reduction Figure 11. Meta-analysis of Exercise Trials in Children: LDL-C Reduction Figure 12. Meta-analysis of Exercise Trials in Children: HDL-C Increase
Tables Table 1. Descriptions of Diets Table 2. Lipid Research Clinics Lipid Levels in U.S. Children and Adolescents Table 3. Total Cholesterol Levels by Race Table 4a. Guideline Definitions of High Risk Table 4b. Inclusion Criteria for Studies of Children with Familial Hypercholesterolemia Table 5a. Cholesterol Tracking by Risk Group at Baseline Table 5b. Cholesterol Tracking by Risk Group at Follow-up Table 6. Summary of Studies Evaluating Sensitivity and Specificity of Family History Table 7a. Summary of Risk Factors for Elevated Total Cholesterol in Children and Adolescents Table 7b. Summary of Risk Factors for Elevated LDL in Children and Adolescents Table 7c. Summary of Risk Factors for Decreased HDL in Children and Adolescents Table 8. Randomized Controlled Trials of Drug Treatment for Children with Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Table 9. Randomized Controlled Trials of Diet Treatment for Children with Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Table 10. Randomized Controlled Trials of Diet Treatment for Children without Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Table 11. Randomized Controlled Trials of Exercise Trials for Children and Adolescents without Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia vii
Table 12. Randomized Controlled Trials of Diet and Exercise for Children without Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Table 13. Adverse Events Reported in Studies of Statins Table 14. Adverse Events Reported in Studies of Bile-Acid Binding Resins Table 15. Adverse Events Reported in Studies of Other Drugs and Combinations Table 16. Adverse Events Reported in Studies of Low Fat Diet Table 17. Adverse Events Reported in Studies of Dietary Supplements Table 18. Summary of Systematic Evidence Review
Evidence Tables Evidence Table 1. Tracking of Serum Lipid Levels Evidence Table 2. Risk Factors Evidence Table 3. Screening Strategies Evidence Table 4. Randomized Controlled Trials of Treatment Evidence Table 5. Adverse Effects of Interventions
Appendixes Appendix 1. Abbreviations Appendix 2. Units of Measure Conversion Formulas Appendix 3. Search Strategies for Child and Adolescent Dyslipidemia Appendix 4. Inclusion/Exclusion Criteria and Results of Searches Appendix 5. USPSTF Quality Rating Criteria Appendix 6. Statistical Methods Used for Meta Analysis
viii
I. INTRODUCTION
Dyslipidemias are disorders of lipoprotein metabolism, including elevations in total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), or triglycerides (TG) or deficiencies of high-density lipoprotein cholesterol (HDL-C) (list of abbreviations in Appendix 1).1, 2 These disorders can be acquired or familial and are related to genetic conditions such as familial hypercholesterolemia in some individuals. Children and adolescents with dyslipidemia may have dyslipidemia as adults.3 The relationship between dyslipidemia and coronary heart disease (CHD) in adults is well-established. Young men (mean age 22) with TC above the 75th percentile followed for 30 years on average, had a relative risk of 2.01 (95% CI 1.59, 2.53) for incidence of CHD.4
The prevalence of other CHD risk factors, such as overweight, diabetes, and metabolic syndrome, is increasing among children and adolescents.5, 6 Using criteria of body mass index (BMI) > 95th percentile, 10% of 2-5 year-olds and 16% of children age 6 years and older are overweight, with higher prevalence in minority racial/ethnic groups.7 Overweight is the primary factor contributing to development of metabolic syndrome in children and adolescents.8 Metabolic syndrome can be defined by glucose level >110 mg/dL, systolic or diastolic blood pressure >90th percentile, TG >110 mg/dL, HDL-C < 40 mg/dL, sex-specific waist circumference >90th percentile. (Please see Appendix 2 for units of measure conversion formulas.) Children and adolescents with metabolic syndrome are more likely to have evidence of inflammation measured by C-reactive protein (CRP) >3 mg/dL.9 Metabolic risk factors are common. In a study of 758 U.S. high school students, 17% had three or more of the following metabolic risk factors: Waist circumference, high triglycerides, LDL-C, fibrinogen, HbgA1C, glucose, insulin and cortisol, and low HDL-C.10
The relationship between childhood and adult dyslipidemia, increasing prevalence of related CHD risk factors in children,5, 11, 12 as well as continued emphasis on primary prevention of CHD has raised interest in screening for dyslipidemia in children. Identification of children with dyslipidemia could lead to intervention services or treatment that might prevent or delay adult dyslipidemia and CHD. This rationale lends support to consideration of screening for dyslipidemia as part of routine well-child care and at other opportunities.
This evidence synthesis focuses on the strengths and limitations of evidence about identifying and managing children and adolescents found to have dyslipidemia by screening in the course of routine primary care. Its objective is to determine the balance of potential benefits and adverse effects of screening for the development of guidelines by the U.S. Preventive Services Task Force (USPSTF). The target population includes children and adolescents age 0-21 years without previously known conditions associated with dyslipidemia. Among this population, there is potential to identify children and adolescents with dyslipidemia from among three groups: Those with undiagnosed monogenic dyslipidemias such as familial hypercholesterolemia (FH), those with undiagnosed secondary causes of dyslipidemia, and those with idiopathic dyslipidemia (polygenetic, multi-factorial or risk factor associated) (Figure 1). This evidence synthesis
1 emphasizes the patient’s or parents’ perspective in the choice of outcome measures and potential adverse effects, and focuses on tests and interventions that are easily interpreted in the context of primary care. It also considers the generalizability of efficacy studies performed in controlled settings and interprets the use of the tests and interventions in community-based populations.
Other systematic evidence reports that contribute information on this topic include Screening and Interventions for Childhood Overweight7 and Screening for Dyslipidemia in Adults: Brief Update of 2001 USPSTF Review.13
Burden of Condition/Epidemiology
Dyslipidemia rarely leads to frank illness in childhood and health effects are typically delayed many years. Lipid levels are low at birth, increasing to young-adult levels by age 2 or 3.14, 15 Although no long-term studies of the direct relationship between blood cholesterol levels measured in children and CHD later in life have been conducted, the relationship between childhood cholesterol levels and CHD can be inferred from indirect evidence. The Muscatine Coronary Risk Factor Survey from 1971-73 measured lipid levels of 2,874 school aged children.16 Children’s cholesterol levels correlated with those of family members, and identification of hypercholesterolemia in children also identified families at risk for CHD.16
The Bogalusa Heart Study is a long-term epidemiologic study of risk factors for CHD from birth through 31 years.17 Seven surveys, including more than 3,500 children, have been conducted since 1973.18 Children of parents with CHD had a higher prevalence of dyslipidemia in childhood.19 Also, there was a correlation between pre-morbid lipid levels and arterial fat deposition by autopsy of children and young adults who died accidentally.20-23 Early lesions of atherosclerosis (fatty streaks) began in the abdominal aorta at age three years, the coronary arteries at age 10 years, and progressed over time.20- 24 The Pathobiological Determinants of Atherosclerosis in Youth (PDAY) study of adolescent males who had died accidentally found atherosclerotic changes of a magnitude directly related to postmortem LDL-C.25, 26
The prevalence of dyslipidemia is determined by the results of laboratory testing and statistically determined criteria. Elevated LDL-C is the most common manifestation of clinically significant dyslipidemia in children. If the cutoff for elevated LDL-C is set at the 95th percentile, then 5% of children in the reference population will have dyslipidemia. Within this group of children with LDL-C levels above the 95th percentile, a minority will have monogenic or secondary dyslipidemias, and the majority will have idiopathic (polygenic, multi-factorial or risk factor associated) dyslipidemias. The more common genetic dyslipidemias are described briefly below. Additional very rare dyslipidemias that can cause adverse effects on health during childhood include homozygous familial hypercholesterolemia, lipoprotein lipase deficiency, Tangier disease, Fish-Eye disease, sitosterolemia, familial hypoalphalipoproteinemia, and abetalipoproteinemia among others.
2 Familial hypercholesterolemia
Familial hypercholesterolemia (FH) results from a defect in the cell surface receptor that removes LDL-C particles from plasma.27 The incidence is 1 in 500 in the U.S., Canada, and Europe.28 and up to 1 in 100 in specific populations (e.g. Transvaal Afrikaners).29, 30 On average, untreated men with FH have clinically evident CHD by age 40 and women by age 50. Most children with FH require drug treatment. The homozygous form of FH can result in TC levels > 500 mg/dL, LDL-C > 450 mg/dL, xanthomas by age 5, and vascular disease before age 20.28, 31 Individuals with heterozygous FH display a broad range of lipid levels, with mean TC of 323+44 mg/dL and mean LDL-C of 262+45mg/dL.32
Familial combined hyperlipidemia
Familial combined hyperlipidemia (FCH) is the most common genetic lipid disorder overall, occurring in 1-2% of the general adult population, and accounts for at least 10% of persons with premature CHD.32 This group tends to have high triglyceride levels, but LDL may be acceptable, borderline, or high (mean 149+48 mg/dL),28 Elevations in LDL, when present, are not generally as extreme as those seen in FH.
Familial defective apoprotein B (apo B)
Familial defective Apo B is characterized by moderate to severe hypercholesterolemia. The mechanism for the effect is a single base substitution in the ApoB gene that diminishes the ability of LDL-C to bind to the LDL-C receptor. Heterozygous familial defective Apo B occurs in 1 in 1,000 Europeans, and approximately 1 in 20 of these have similar phenotypes to those with heterozygous FH.28
Familial hypertriglyceridemia
Familial hypertriglyceridemia is autosomal dominant but usually not expressed until adulthood. Obesity can accelerate expression of the phenotype, with moderately elevated (100-200 mg/dL) triglyceride levels in youth and extremely elevated levels in adults.28
In a study of diagnoses made systematically for 129 families referred to a single lipid clinic for dyslipidemia and family history of early CHD, 20 had FH, 65 had FCH, 11 had hyper-Apo B and 1 had familial hypertriglyceridemia33. Others were unexplained (32), presumed normal (17) or adopted (2).
In addition to detection of monogenic dyslipidemias, screening would also detect secondary causes of dyslipidemia such as diabetes, nephritic syndrome, hypothyroidism and others, and idiopathic dyslipidemia. Included in the idiopathic group are polygenic, environmental and behavioral causes such as overweight and smoking. While children
3 and adolescents with idiopathic dyslipidemia generally have less severe lipid levels than children and adolescents with familial or rare genetic disorders, levels may be significantly abnormal and could potentially improve with intervention.
Healthcare Interventions
To reduce the burden of dyslipidemia in children, clinic-based screening, community- based screening, and other community-based prevention strategies have been proposed. Most recommendations support selective strategies that test children who have family members with dyslipidemia or premature CHD and those for whom family history is unknown.34, 35 Alternatively, universal screening for all children has had some proponents, although this approach has not been recommended in recent guidelines.
For children over two years of age with dyslipidemia, most recommendations indicate that the initial intervention is a low-fat, low-cholesterol diet, such as the American Heart Association Step I diet (AHA)11 (Table 1). Children younger than two years should not be prescribed a low fat diet because the first two years is a period of rapid growth and development that requires an adequate fat and cholesterol intake.35 If the AHA Step I diet alone does not result in satisfactory improvement in the lipid profile, the AHA Step II diet may be prescribed. This is a more restrictive diet, lower still in fat and cholesterol content than the Step I diet, and should not be prescribed to children without close supervision by a physician and dietitian. If the Step II diet still does not lead to a satisfactory improvement, medications can be considered. In 2000, the AHA modified their dietary recommendations for children,36 but these changes have not been reflected in American Academy of Pediatrics (AAP) or National Cholesterol Education Program (NCEP) guidelines.
Children and adolescents with FH are the only non-adults for whom trials of drug therapy are available and approved by the U.S. Food and Drug Administration (FDA). The only medications approved by the FDA for treatment of dyslipidemia for children younger than 8 years of age are bile acid-binding resins. However, adherence to a prescribed regimen of bile acid binding resins is difficult due to the character and taste of the compounds. In addition, the lipid lowering effects of bile acid-binding resins in children are limited and do not often result in lipid lowering to the degree desired. Recently, HMG Co-A reductase inhibitors (statins) were approved for use in children with heterozygous familial hypercholesterolemia (FH). Lovastatin, simvastatin, and atorvastatin are FDA approved for adolescent boys and girls at least one year post- menarche ages 10-17, and pravastatin for boys and girls ages 8-18.37, 38 Other medications used in adults for treatment of hyperlipidemia are either not recommended for children (i.e., niacin) or have not been adequately evaluated for safety and efficacy in children.
Additional interventions often recommended in the management of dyslipidemia in children and adolescents include dietary counseling, exercise, weight loss for overweight children, identification and treatment of diabetes mellitus or other secondary cause, and
4 control of high blood pressure. Dietary supplements such as fiber, omega 3 fatty acids, and sterol or stanol margarines are sometimes considered as interventions for children with dyslipidemia.39-42
Prior Recommendations
In 2001, the USPSTF reviewed screening for dyslipidemia in children and adolescents but did not make a recommendation.43 A 2001 Systematic Evidence Review for the USPSTF found that studies of drug therapy in children were too short (8 weeks to 1 year) and too small to draw definitive conclusions about harms or benefits. Determination of the efficacy, safety, and feasibility of low fat diets in children were also inconclusive.44
The NCEP recommendations for adults (Adult Treatment Panel III [ATP III]) recommended that screening begin at age 20 and were updated as recently as 2004.45 The ATP III and USPSTF guidelines for adults recommend initial testing and risk determination using TC and HDL-C.13, 34 LDL-C is used as a criterion for initiation of drug therapy.34
The NCEP Report of the Expert Panel on Blood Cholesterol Levels in Children and Adolescents published guidelines in 1992 and have not been updated.34 This report addressed children up to age 18, and is the basis for the most recent AAP and AHA guidelines. The NCEP report recommended selective screening for children and adolescents with a family history of premature CHD or at least one parent with high TC (TC > 240 mg/dL) in the context of regular health care. It recommended that cardiologists make a routine practice of referring the offspring of their adult patients with premature CHD to a source of continuing health care for cholesterol testing and follow- up. Optional cholesterol testing may be appropriate in children and adolescents judged to be at higher risk independent of family history or parental hypercholesterolemia. For example, children and adolescents who are overweight and/or consume excessive amounts of saturated fatty acids, total fat, and/or cholesterol may warrant testing.
The NCEP screening protocol for children and adolescents varies according to the reason for testing (Figure 2). For those being tested because of parental hypercholesterolemia, NCEP guidelines suggest that a non-fasting total cholesterol be the initial test. Further testing with a lipoprotein profile is recommended depending on the level of total cholesterol. For those being tested for family history of premature CHD, a fasting lipoprotein analysis is recommended.34
Interventions are determined by results of the fasting lipoprotein analysis (Figure 2). The NCEP panel recommended considering drug therapy in children aged 10 years and older if, after an adequate trial of diet therapy (6-12 months), LDL-C cholesterol remains >190 mg/dL, or LDL-C cholesterol remains >160 mg/dL and there is a family history of premature CHD or at least two or more other risk factors are present. Referral to a specialized lipid center may be appropriate in some cases, especially when high
5 cholesterol is due to secondary causes or accompanied by multiple risk factors and a family history of premature CHD.34
In preparing these recommendations, the NCEP decided not to recommend universal screening for the following reasons: 1) Although high cholesterol levels in childhood generally predict cholesterol elevations in adulthood, many children with high cholesterol levels will not have high enough levels as adults to qualify for individualized treatment; 2) Universal screening could lead to the labeling of many young people as patients with a “disease,” causing unjustified anxiety for them and their families; 3) For most children not from high-risk families, there is sufficient opportunity to begin cholesterol-lowering therapies when they reach adulthood; 4) There is insufficient evidence concerning the long-term safety and efficacy of drug therapy in childhood to reduce CHD morbidity and mortality in adulthood and universal screening could lead to overuse of cholesterol- lowering drugs in childhood and adolescence.34, 46
The American Academy of Pediatrics Committee on Nutrition published additional guidelines in 1998, slightly modifying the NCEP Report.35 The committee noted that although the precise fraction of risk for future CHD conveyed by elevated cholesterol levels in childhood is unknown, clear epidemiologic and experimental evidence indicates that the risk is significant. The guidelines stated that diet changes that lower total fat, saturated fat, and cholesterol intake in children > age 2 and adolescents can be applied safely and acceptably, resulting in improved plasma lipid profiles that, if carried into adult life, have the potential to reduce atherosclerotic vascular disease. The guideline also recommended reducing other risk factors. There was no change in the recommended initial screening tests.
The American Heart Association’s (AHA) initial “Guide to the Primary Prevention of Cardiovascular Disease,” published in 1996 and updated in 2002, does not address prevention in children.11 In 2003, the AHA published updated guidelines for children that generally follow those set forth by the NCEP Pediatric Panel in 1992.11 The new AHA guidelines suggest that bile acid-binding resins or statins are usual first-line agents in the treatment of severe dyslipidemia, and that pharmacologic intervention for dyslipidemia should be accomplished in collaboration with a physician experienced in treatment of disorders of cholesterol in pediatric patients. This report also considered hypertriglyceridemia and low HDL-C, with the recommendation of a goal of a fasting triglyceride level <150 mg/dL, and HDL-C >35 mg/dL. No pharmacologic interventions were recommended in children with isolated elevation of fasting triglycerides unless >400 mg/dL, at which level there is an increased risk of pancreatitis.11
Scope of Evidence Synthesis
The patient population, interventions, outcomes, and adverse effects of screening and treatment are summarized in an analytic framework (Figure 3). Corresponding key questions guided the literature review and evidence synthesis. The key questions examine a chain of evidence about the accuracy and feasibility of screening children and
6 adolescents for dyslipidemia in primary care or community settings, tracking of lipid levels through childhood to adulthood, role of risk factors in selecting children and adolescents for screening, effectiveness of interventions for children and adolescents identified with dyslipidemia, and adverse effects of screening and interventions. This review includes treatment trials of children and adolescents using dietary, exercise, and drug interventions. Studies of children with previously diagnosed conditions known to cause dyslipidemia (e.g. secondary dyslipidemias and monogenic dyslipidemias) were not included because the scope of this review is screening children without known diagnoses. Specifically, studies of children with diabetes were not included because these children would ordinarily already be under surveillance for lipid disorders as a result of their primary disease. Trials of drug therapy in children with heterozygous FH or FCH are included because this population has been exclusively enrolled in treatment trials. Data on cholesterol levels and diagnostic criteria for heterozygous FH are included because identification of previously undiagnosed FH is also a goal of screening (Figure 1).
7 II. METHODS
Literature Search and Strategy
Relevant studies were identified from multiple searches of MEDLINE (1966 through September 2005) [Appendix 3]. Additional articles were obtained from recent systematic reviews, reference lists of related studies, reviews, editorials, websites and by consulting experts. Retrieved abstracts were entered into an electronic database (EndNote®).
Inclusion/Exclusion Criteria
Investigators reviewed all abstracts identified by the searches and determined eligibility by applying inclusion and exclusion criteria specific to each key question (Appendix 4). Full-text articles of included abstracts were then reviewed for relevance. Eligible studies were English-language, applicable to U.S. clinical practice, and provided primary data relevant to key questions. Studies of children and adolescents with previously diagnosed conditions known to cause dyslipidemia were not included. Animal studies were not included. Studies of risk factors were included only if they provided multivariate adjusted analyses.
For treatment studies, full text randomized controlled trials (RCTs), non-controlled clinical trials, and non-controlled prospective studies providing data on the treatment of children and adolescents with diet, drug therapy, exercise, or combinations of these were initially reviewed. Subsequently, only randomized controlled trials and meta-analyses of randomized controlled trials that reported serum lipid outcomes were included. For Key Question 10, outcomes included either adult lipid levels or adult CHD. Studies of other treatment modalities for dyslipidemia (surgery, apheresis) and studies of rare conditions such as homozygous FH were excluded. Information about adverse effects of treatment was obtained from RCTs and additional sources such as non-randomized controlled treatment trials and non-comparative studies of treatment.
Data Extraction and Synthesis
All eligible studies were reviewed and a “best evidence” approach was applied, in which studies with the highest quality and most rigorous design are emphasized.47 Data were extracted from each study, entered directly into evidence tables, and summarized. Benefits and adverse effects of therapies were considered equally important and both types of outcomes were abstracted. Trials of therapy for children and adolescents with dyslipidemia were categorized by population and intervention.
8 Two reviewers independently rated the quality of randomized controlled trials using criteria specific to different study designs developed by the U.S. Preventive Services Task Force (Appendix 5).48 The overall rating is a combination of internal and external validity scores. When reviewers disagreed, a final rating was reached through consensus.
Randomized controlled trials of similar treatments that met additional eligibility criteria were considered for meta-analysis. Meta-analyses were performed to provide estimates of the effectiveness of statins on improving lipid levels in children and adolescents with FH, and of the effectiveness of exercise on improving lipid levels in children and adolescents without FH who were normal or overweight. For each trial, the difference in mean percent change of lipid levels between treatment and control groups and its standard error were obtained and pooled using a random effects model.49 When the percentage change and its standard error or 95% confidence interval were not reported, they were calculated from the mean and standard error of lipid levels from treatment and control groups at baseline and the endpoint (Appendix 6). A study was excluded when no information on dispersion was reported.
Effects of study level covariates, such as duration, mean age, percentage male/female, and dosage were checked by using random-effects meta regressions.49 Specifically, drug dose for each statin study was analyzed using an equivalent dose of simvastatin according to published equivalency tables.50
Size of Literature Reviewed
A total of 2,507 unique citations were identified by the literature searches (Appendix 4). Included were 160 papers about screening and testing for dyslipidemia (Key Question 2); 68 about interventions and tracking of lipid values over time (Key Questions 4-8 and 10); 8 about the adverse effects of screening (Key Question 3); and 81 about adverse effects of treatment (Key Question 9). Seven papers discussed the costs of screening, but none evaluated cost effectiveness in U.S. population (Key Question 11).
External Review Process
The USPSTF appointed liaisons to advise the Oregon Evidence-based Practice Center in formulating and reporting this systematic evidence review. An additional set of outside experts provided advice in the review formulation stage and commented on a draft version of the evidence synthesis.
9 III. RESULTS
Key Question 1. Is Screening for Dyslipidemia in Children/adolescents Effective in Delaying the Onset and Reducing the Incidence of CHD-related Events?
No studies evaluated the effect of screening children and adolescents on adult lipid or disease outcomes.
Key Question 2. What is the Accuracy of Screening for Dyslipidemia in Identifying Children/adolescents at Increased Risk of CHD-related Events and Other Outcomes?
The following sub-questions to Key Question 2 address the accuracy and feasibility of screening children for dyslipidemia.
Key Question 2a. What are Abnormal Lipid Values in Children/adolescents?
Summary
Normal lipid values for children are defined according to percentile cut points within population distributions. In large population-based samples, TC levels increase from birth, stabilize at approximately age two years, peak prior to puberty, then decline slightly. NCEP recommendations are based on results from the Lipid Research Clinics (LRC) Prevalence Study, which determined the 95th percentile cut point for TC as 200 mg/dL and for LDL-C as 130 mg/dL. More recent studies indicate age, sex, and racial differences and temporal trends that shift these cut points.
Evidence
Although several studies conducted in the U.S. during the 1970s obtained lipid levels from large samples of normal healthy children,51-53 current recommendations11, 34, 35, 54 are based on distributions of lipid and lipoprotein levels obtained from the Lipid Research Clinics (LRC) Prevalence Study.14 This study included one Canadian and nine U.S. sites and enrolled subjects primarily based on residency within census tracts, school enrollment, and employment in occupational and industrial groups. Fasting (≥12 hours) lipid and lipoprotein levels were obtained in 15,626 children age 0-19 years between 1972-1976. The selected populations included a broad range of geographic, socio-
10 economic, occupational, sex, and ethnic groups, but they were not selected to be a representative sample of the North American population.
In the LRC sample, TC levels increased from birth and stabilized at approximately age two years. At puberty, TC levels declined slightly for both boys and girls, and HDL-C levels declined for boys. For all children, the mean serum level for TC was approximately 160 mg/dL and for LDL-C was 100 mg/dL. The 95th percentile level was 200 mg/dL for TC and 130 mg/dL for LDL-C (Table 2). Results for African American children were similar, however, they were based on smaller numbers and provided TC and triglycerides only.14
To determine racial differences, a study of a multi-racial sample of 6,585 U.S. children ages 5-18 years from 22 schools in five states reported non-fasting finger-stick plasma TC levels from 1984-88.55 The mean TC level for African American children (173 mg/dL) was significantly higher (p<0.001) than for Hispanics (168 mg/dL), Asians (165 mg/dL), and Whites (163 mg/dL) (Table 3). The mean TC level for Hispanics was significantly higher than for Whites (p<0.001) but not Asians, and higher for girls than boys for Whites only (165 mg/dL vs. 160 mg/dL, p<0.001).
More recent data from the National Health and Nutrition Examination Survey (NHANES) III (1988-94) confirm many of these findings. However, in NHANES III mean TC levels among 12-17 year olds decreased by 7 mg/dL from the National Health Examination Survey (NHES) III (1966-1970) and NHANES I (1971-1974) levels shifting percentile cut points upward.56 NHANES III data from 7,499 children and adolescents ages 4-19 years indicate a 95th percentile cut point for serum TC as 216 mg/dL and for LDL-C as 152 mg/dL.56 Mean age-specific TC levels peaked at 171 mg/dL at 9-11 years and declined at older ages. Girls had significantly higher mean TC and LDL-C levels than boys (p<0.005). Non-Hispanic Black children and adolescents had significantly higher mean TC, LDL-C, and HDL-C levels compared to non-Hispanic White and Mexican American children and adolescents. In linear regression models of these data, age, sex, and race have significant effects on lipid levels questioning the utility of fixed screening cut points.57
Key Question 2b. What are the Appropriate Tests? How Well do Screening Tests (Non-fasting Total Cholesterol, Fasting Total Cholesterol, Fasting Lipoprotein Analysis) Identify Individuals with Dyslipidemia?
An appropriate screening test is one that has a strong association with future risk for CHD, or one that can be used to identify people who benefit from treatment.
11 Summary
TC minus HDL-C above the 95th percentile is 88-96% sensitive and 98% specific for detecting LDL-C >130 mg/dL. However, a single TC or HDL-C measurement is insufficient for practitioners to assign children and adolescents to the NCEP risk categories with 95% confidence.
Identification of familial hypercholesterolemia (FH) has traditionally used LDL-C levels alone or combined with family history. In families with known FH, LDL-C levels above the 95th percentile detect FH among first-degree relatives with 95% sensitivity. However, the ability of LDL-C to predict FH decreases when using second and third degree relatives of an FH proband, and decreases further when applying these criteria to the general population. Lipid Research Clinics (LRC) data indicate that TC > the 95th percentile is 69% sensitive and 98% specific for detecting children and adolescents (ages 6-19) with LDL-C > the 95th percentile and TG < the 95th percentile for age and gender. No studies have evaluated lipid levels in the general U.S. population as a diagnostic test for FH using mutation or biochemical analysis as the gold standard.
Evidence
In AAP and NCEP guidelines, TC is used as an initial laboratory measurement for children being tested because of a family history of high cholesterol, and a lipoprotein profile is obtained if the patient has a TC over a certain defined target. In children, as in adults, LDL-C is the basis for initiating treatment and determining goals of therapy. How well TC levels detect elevated LDL-C levels has been examined in three different populations, using LRC data (ages 6-19, n=1325),58 Hispanic children (ages 4-5, n=106)59 and the biracial Bogalusa cohort (ages 5-17, n=2,857).60 TC> the 95th percentile detected LDL > the 95th percentile with sensitivities of 44% (White females) to 50% (White males, African American males and females).60 Specificities were 90% (African American and White males and females).60 With LRC data, a fasting TC > the 95th percentile had 69% sensitivity and 98% specificity for detecting children with LDL > the 95th percentile and TG
In adults, the recommended screening tests are TC and HDL-C. While this has not been recommended in guidelines for children and adolescents, it is common in practice.61 HDL-C may help distinguish false negatives from true negatives among those screened with TC alone.58 The use of TC and HDL-C in combination as a screening tool has been evaluated62, 63 In 260 African American adolescents (ages 12-20), fasting TC minus
12 HDL-C above the 95th percentile was 96% sensitive and 98% specific for predicting LDL-C >130 mg/dL. Using data from adolescents who were both non-fasting and fasting, TC minus HDL-C was 88% sensitive and 98% specific.62
Using weekly repeated measures of TC and HDL-C, within-person variability and the ability of these tests to reliably classify children and adolescents into NCEP risk categories (TC<170 mg/dL acceptable, TC 170-199 mg/dL borderline, and >200 mg/dL high; and for HDL-C <35 mg/dL low, and >35 mg/dL acceptable) were assessed. A child needed to have a TC level below 155 mg/dL or above 185 mg/dL to be assigned a category of acceptable or borderline/high (respectively) based on a single measurement. To accurately assign a child to the high category with 95% confidence, a TC > 215 mg/dL was required. No single TC value would place a child in the borderline category (170-200 mg/dL) with 95% confidence. Averaging two TC readings improved the ability to classify children with 95% confidence: a TC <159 mg/dL could reliably assign a child to the acceptable category and a TC of >211 mg/dL could reliably assign a child to the high category. Three TC measurements did not substantially improve the ability to classify children into risk categories.63
Likewise, one measurement of HDL-C was required to be <28 mg/dL or > 43 mg/dL to confidently assign (based on the 95% confidence interval) the child to a low HDL (<35 mg/dL) or acceptable (>35 mg/dL) category. With two HDL measurements, the values required were <30 mg/dL and >40 mg/dL respectively.63
Direct LDL-C measurement offers the advantage of using nonfasting serum samples. It may be as precise as that calculated LDL-C using the Friedewald formula,64 but this remains controversial. Immunoseparation and direct determination of plasma LDL-C correctly classified 81% of children according to NCEP cut-offs, compared to 84% of children correctly classified using the Friedewald formula.64 In another study, however, direct LDL-C correctly classified 83% and 81% of those children with LDL-C >130 mg/dL and >190 mg/dL respectively, but only 43-55% of those with LDL-C 130-190 mg/dL. This compared with the Friedwald formula which correctly classified 79-82% of the group between 130-190 (using a gold standard that was β-quantification LDL).65
A primary goal of screening in children and adolescents is to identify individuals who are at risk for premature CHD and would benefit from early treatment. The most prevalent condition leading to early CHD is familial hypercholesterolemia (FH). Clinical definitions of FH vary (Table 4A34-36 and 4B38, 39, 66-83). Many studies assume FH is present based on LDL-C levels and clinical data. Studies have used LDL-C levels above the 95th percentile for age and gender with a family history of early CHD as a diagnostic criterion for FH. Others have used LDL-C above a certain level (LDL-C cutoff varies among studies) and a parent with FH, but it is often unclear how the parental diagnosis was made. Still others have used LDL-C levels along with the presence of xanthomas to define FH, though xanthomas are rare in children with FH, especially young children.
The gold standards for diagnosis of FH include either molecular genetic studies identifying a pathogenic mutation in the LDL-C receptor gene, or biochemical studies
13 showing defective LDL-C receptor activity in cultured fibroblasts. Over 800 LDL-C receptor mutations have been identified in FH,30, 84, 85 thereby complicating diagnostic confirmation. Some populations (South African Afrikaners, Finnish, French Canadian) have limited numbers of mutations,30, 85 but this is not true of the U.S. population. LDL- C receptor activity assays are expensive, difficult and not widely available.30 As a result, clinical trials reporting results in children with FH for the most part included children who likely have FH but who have not had diagnostic confirmation (Table 4B). Although genetic screening has been evaluated in populations where there are few known mutations causing FH,86, 87 no studies have evaluated LDL-C levels in a general population of U.S. children identified through lipid screening and confirmed suspected cases of FH with mutation analysis. Several descriptive epidemiologic studies have used LDL-C levels for diagnosis within identified FH families88-90 Even among these studies, only one used families identified by mutation analysis,89 whereas the others used families identified by either mutation analysis or LDL-C above the 95th percentile along with tendon xanthomata,88 or by lipid levels alone.29 LDL-C levels greater than 135 mg/dL,88 and LDL-C levels above the 95th percentile for age and gender among 1st degree relatives of an FH proband89 captured 95% of FH cases within families.
The difference in the predictive value of LDL-C levels for diagnosing individuals within families as opposed to in the general population was nicely demonstrated by a study of relatives from 5 large, LDL-C receptor mutation confirmed Utah FH pedigrees. Data from this cohort were used to assess the sensitivity and specificity of mathematically- derived screening criteria in the population of relatives.91 Differences in predictive values for total cholesterol among 1st, 2nd, and 3rd degree relatives of persons with FH were evaluated and demonstrated graphically (Figure 4). The predictive value of a specific LDL-C level to diagnose FH decreases substantially for 1st, 2nd, and then 3rd degree relatives of FH, and is even lower for the general population. Thus, TC and LDL-C levels that were evaluated for diagnosing FH using FH kindreds do not allow for application of identified LDL-C levels to non-FH families or to general populations. There were no studies identified that evaluated LDL-C levels in a general population of children or adolescents that subsequently confirmed suspected (screen identified) cases of FH with mutation or biochemical analysis.
Key Question 2c. How Well do Lipid Levels Track from Childhood to Adulthood?
Summary
Serial correlations between lipid levels measured in individual children over time are higher for TC (r=0.38-0.78) and LDL-C (r=0.4-0.7), than for HDL-C (r=0-0.8) and TG (r=0.1-0.58). Tracking is generally consistent within a cohort. Age at first measurement and duration of follow-up do not appear to be related to strength of the correlation between measurements. Approximately 40-55% of children with elevated lipids defined
14 by percentile within a population distribution will continue to have elevated lipids on follow-up. Likewise, the highest quintile of TC at follow-up consists of children and adolescents from the 1st (4-6%), the 2nd (11-13%), the 3rd (19-20%) and the 4th (18-21%) quintiles at baseline. There were no studies of tracking according to other risk factors.
Evidence
Twenty-three prospective cohort studies contributed information on tracking of lipid levels during childhood. Twenty-one studies reported correlation coefficients and are listed on Evidence Table 1,3, 17, 92-110 whereas 2 studies reported tracking only by percentile and are described in the text.111, 112 These studies drew from seven U.S. cohorts and eight non-U.S. cohorts.
Correlation coefficients from these studies are displayed both by age at first measurement (Figures 5A-5D) and duration of follow-up (Figures 6A-6D). In both displays, TC and LDL-C track more consistently than do HDL-C and TG. There does not appear to be a clear pattern of improvement in tracking of cholesterol levels with either increasing age or shorter duration of follow-up. Studies noted better tracking for boys and African Americans.3, 17, 108 The range of correlation coefficients (or tracking coefficients) among the U.S. cohort studies is 0.38 to 0.78 for TC, 0.4 to 0.71 for LDL-C, 0 to 0.79 for HDL- C and 0.1 to 0.58 for TG. HDL-C tracking was worse in one study from the Muscatine cohort (0 to 0.1) than in studies from other cohorts (0.2 to 0.79). Studies also reported the percentage of children remaining in high-cholesterol classification categories at follow-up (Table 5A).3, 92, 94, 95, 98, 99, 105, 111 Of children ages 5- 18 with two consecutive cholesterol measurements above the 90th percentile in the Muscatine cohort who had lipids measured again at ages 20-30 (n=70), 34% had adult cholesterol levels > 240 mg/dL, 37% had adult levels between 200-239 mg/dL, and 29% had adult levels < 200 mg/dL.113 In the Bogalusa cohort, 196 children were classified in the highest quintile of TC at year one; of those, 55% remained in the highest quintile eight years later.17 The Child and Adolescent Trial for Cardiovascular Health (CATCH) cohort study is large (n=3,659) and more recent, but has not yet followed the cohort through to adulthood.3 Between 3rd and 8th grades, 53% remained in the lower quintile and 55% in the highest quintile of TC. Fifty-three percent remained in the lowest quintile and 55% in the highest quintile for HDL-C.3
Likewise, some studies have evaluated the derivation of the highest risk group at follow- up (Table 5B).96, 105, 108, 112 In the Beaver County Lipid Study individuals who were in the top quintile for TC at follow-up (ages 17-30 years), had the following distribution of TC values at baseline (ages 1-14 years): 6% had TC in the 1st (lowest) quintile, 13% in the 2nd, 19% in the 3rd, 21% in the 4th, and 40% in the 5th (highest).96
Most tracking studies compared single lipid measurements over time. Repeated lipid measurements in children show variation (see above, Key Question 2a). Therefore, regression to the mean should be considered as one explanation for variability among tracked values.108, 114 Some adjusted for covariates, or used tracking coefficients.101
15 None of these studies evaluated the proportion of children and adolescents with lipid levels above the 95th percentile who remained in the top 5% at follow-up.
Key Question 2d. What is the Accuracy of Family History in Determining Risk?
Summary
Family histories can be time consuming and difficult to elicit accurately. Screening using family history has been shown to miss substantial numbers (30-60% in general) of children with elevated lipids. Family history definitions vary substantially among studies as do lipid detection thresholds. Those studies that show higher sensitivities (~77%) have low specificities (55% or lower). Population-based estimates of the number of children requiring serum lipid testing based on family history may range from 25-55%, depending on the exact definition of family history and the lipid detection threshold.
Evidence
The AAP and NCEP advocate selective screening for dyslipidemia among children and adolescents who have a family history of early CHD or high cholesterol. Many good quality studies of diagnostic accuracy have evaluated the sensitivity and specificity of family history according to various guidelines for detecting abnormal lipids in children and adolescents (Table 6).59, 62, 115-128 All of these studies used an appropriately broad spectrum of patients either from pediatricians’ offices, schools, or ongoing cohort studies and applied tests of family history and lipid measurement equally. The studies used different definitions of the screening tool (any parental history of heart attack, other parental risk factors, varied age definitions of early CHD), and selected different levels of LDL-C or TC as the lipid detection threshold.
Definitions of a positive family history can vary, including whether grandparents are considered, the age used to define premature CHD (55, 60, or 65, or gender-based), and whether CHD in 1st and 2nd degree relatives is taken into account. Likewise, in these studies, multiple different definitions of family history were evaluated. A family history definition that included a parent or grandparent with CHD prior to age 50-60 and/or cholesterol >240 mg/dL had sensitivities ranging from 46-74% for TC >170 or LDL-C >130 mg/dL.59, 117 Parental history of early CHD alone was 5-17% sensitive for TC >170 mg/dL or LDL-C >130 mg/dL,59, 117 whereas parent or grandparent history of early CHD was 46% sensitive for LDL >the 95th percentile.119
Regardless of the precise definition, family history has been repeatedly shown to miss substantial numbers of children with elevated lipids, ranging from 17-90% overall and 30-60% in most studies.62, 118, 119, 122, 124, 126, 129-131 In addition, the proportion of children
16 and adolescents who qualify for screening based on family history is generally between 25-55% depending on the sensitivity of the specific family history question.59, 62, 115-119, 121, 124, 125, 127, 132 This compares with the original NCEP estimate based on LRC data that 25% of the child/adolescent population would require screening on the basis of family history.133 Studies reporting sensitivities above 70% used family history of high cholesterol, early CHD (< age 60), stroke, angina, or hypertension to detect a TC >200 mg/dL;115 family history of high cholesterol or early CHD to detect LDL >130 mg/dL117; or parent or grandparent history of any risk factor (high cholesterol, diabetes, hypertension, gout, obesity) or atherosclerotic risk factor (MI, angina, sudden death, stroke or PVD) prior to age 55 to detect a fasting LDL >the 95th percentile.119 Of these studies with the highest sensitivity, most had poor specificity, the highest being 55%.117
Obtaining family history can be time consuming and the history is often inaccurate depending on who is giving it. There is concern that African-American Black children and children from single-parent households are more likely to be missed by screening strategies that use family history.126, 133
Key Question 2e. What are Other Important Risk Factors?
Summary
Evidence for child and adolescent risk factors for dyslipidemia other than family history is strongest for overweight, but otherwise is lacking for the general population. Epidemiologic cross-sectional and cohort studies report a strong statistical association between overweight and elevations in lipids. The potential impact of incorporating weight measures into a screening tool has not been addressed. Multiple other risk factors (diet, physical inactivity, aerobic capacity/fitness, puberty and smoking) have not been evaluated adequately to assess their contribution to dyslipidemia in children or their usefulness as screening tools.
Evidence
In addition to family history, assessment tools for selective screening could incorporate other risk factors reflecting a child’s diet, physical activity level, aerobic capacity/cardiovascular fitness, overweight, and fat distribution. Childhood overweight, as measured by BMI, has been identified in one review as the best independent predictor of adult dyslipidemia after LDL-C. Its predictive power is most notable when considering increase in BMI from childhood to adulthood.134
Forty-two cohort and cross-sectional studies of mixed quality with adjusted statistical analyses contributed information on additional risk factors for identifying children at
17 increased risk for elevated lipids and/or CHD-related events (Evidence Table 2, Tables 7A, 7B, 7C).18, 120, 135-174
Thirty studies examined overweight or body fat composition measures as a risk factor18, 135-141, 144-149, 154, 156-159, 161-167, 169, 170, 172, 174 and it was the most consistently effective in predicting risk of dyslipidemia compared to other factors assessed.
Six studies evaluated overweight as a risk factor along with family history.140, 141, 148, 165, 170, 172 Italian school girls (n=361) who were overweight, defined as weighing > 20% ideal weight, had higher mean LDL-C (126.6 mg/dL for overweight vs 99.5 for those not overweight mg/dL, p<0.001) and mean TC (173.8 mg/dL vs 199.0 mg/dL, p<0.01) with lower mean HDL-C (58.9 mg/dL vs 50.0 mg/dL, p<0.001).148
Among 506 inner-city youths aged 5 to 19 who were predominantly African American and Hispanic, those who were overweight had a 1.74 odds ratio for having hypercholesterolemia controlling for age and family history (95% CI=1.34-2.14).141 Sensitivity of using this overweight measure alone to screen was 42%, with a specificity of 71% for predicting hypercholesterolemia (TC > 170 mg/dL or LDL-C > 110 mg/dL). Combining overweight with family history had a sensitivity of 49% and specificity of 64%.141
A Wisconsin study of 2,726 schoolchildren aged 5 to 15 showed BMI was related to adverse levels of all lipids and lipoproteins (TC: r= 0.010 [95% C.I. 0.002, 0.018], LDL- C: r= 0.012 [95% C.I. 0.004, 0.019], HDL-C: r= -0.013 [95% C.I. -0.016, -0.010]).140 In contrast, a study of 1,081 youths aged 2 to 20 in a pediatric office-based cholesterol screening program showed BMI was not associated with cholesterol level.172 A study of 458 youths in a suburban Dutch community found that for girls over age 15, TC increased step-wise across quartiles of body weight with the 4th quartile showing a significant difference from the 1st quartile of approximately 7mg/100ml (p<0.05).170 A multiple regression analysis of 164 highly-selected youths aged 6–16 with heterozygous FH showed that LDL-C was related to FH parent’s cholesterol levels, percent body fat, pubertal stage, and non-FH parent’s cholesterol level (p<.01), together explaining 40% of variance in LDL-C (CI 25%-55%).165
Key Question 2f. What are Effective Screening Strategies for Children/adolescents (Including Frequency of Testing, Optimal Age for Testing)?
Summary
Currently recommended screening strategies have limited diagnostic accuracy, low adherence to guidelines by providers, and limited compliance by parents and children.
18 No trials compare strategies of screening in children, address the frequency and optimal age for testing, or evaluate strategies that might increase provider and parent compliance.
Evidence
Twenty-six studies evaluated screening strategies among children in various settings (Evidence Table 3).59, 62, 79, 115, 117-120, 122, 124, 126, 130, 131, 133, 175-186 One RCT of college students compared two regimens for screening;183 all others were non-comparative prospective studies describing screening interventions. Two additional studies addressed parent and physician response to screening results.187, 188 These and others reported low parental compliance with follow-up testing, even when follow-up was free, as with pre- paid health plans.
Studies demonstrate low compliance among primary care physicians in following current guidelines for screening.188 In an ancillary study of the Child Adolescent Trial for Cardiovascular Health (CATCH), parents were given recommendations to consult their child’s physician if TC exceeded 200 mg/dL on one or more occasion.187 After physicians saw the children, only 59% were further evaluated for possible elevated cholesterol. Of these, half of the physicians repeated cholesterol tests, 42% asked about family history, 38% made recommendations for dietary management, and only 12% referred children to dietitians.187
Neonatal screening
Neonatal screening for dyslipidemia has been examined in multiple studies, mostly using cord blood for analysis of TC106, 189-202 Other studies have investigated the use of dried filter paper blood spots from cord blood203 and from heel sticks taken from 3-7 day old infants (the same samples used for routine neonatal screening for conditions such as hypothyroidism and phenylketonuria).204-209 Five studies had follow-up of cord blood cholesterol levels with serum levels in the children at a later time.190, 192, 193, 196, 197 One study included serum analysis at birth and 1 year, along with LDL receptor mutation testing of infants with parents who had DNA-documented heterozygous FH, comparing levels in affected newborns to non-affected newborns.210 Using cord blood, TC and LDL- C levels were elevated in the FH newborns as compared to non-FH newborns, but there were no significant differences between HDL-C or TG. However, differences between FH and non-FH infants in TC and LDL-C were more pronounced when measured at 1 year.210 No studies were identified that screened a general population of infants and followed up abnormal results with mutation analysis.
19 Key Question 3. What are the Adverse Effects of Screening (Including False Positives, False Negatives, Labeling)?
Summary
Parental non-compliance with screening and follow-up recommendations is reported to be related to concerns about test accuracy, lack of proof that intervention makes a difference in children, concerns about upsetting the child, refusal by the child, inconvenience, or decisions to institute a diet themselves and have child rechecked subsequently.
Evidence
Potential adverse effects of screening for dyslipidemia among children have been examined in one randomized controlled trial211 and five studies of various other designs130, 212-215 Screened college students were randomized to receive desirable results (TC = 174 mg/dL) or borderline-high results (TC = 224 mg/dL). Those who received borderline-high test results rated high cholesterol as a less serious threat to health than did the desirable-feedback subjects, and perceived the test to be less accurate. Higher self- esteem was associated with lower likelihood of requesting information after receiving borderline-high test results.211
Parental non-compliance with screening has been demonstrated by two studies of screening. The Child and Adolescent Trial for Cardiovascular Health (CATCH) program measured cholesterol levels in 5,106 children in 3rd grade and in 3,936 of these again in 5th grade. Parents received mailed notification of their children’s cholesterol results and a follow-up telephone survey. Parents indicated that a physician had been consulted for only 20% of children. Eighty-four percent of those notified once recalled receiving the notice, yet only 39% of those remembered the results as being high.187
Likewise, a pediatric group screened 439 children ages 2-15 as part of well-child exams or non-acute illness including siblings of all those seen. Of the 134 (31%) children who had cholesterol levels above the 75th percentile and were recommended to return for a fasting lipid panel, 51% did so. Parents of children who did not return were contacted 4- 6 months later, and 94% remembered their child’s screening cholesterol level had been elevated, while 64% remembered that follow-up had been suggested. The most common reason parents provided for lack of timely follow-up were concerns about whether the test was accurate (67%), planning to recheck at next visit (67%), concern that child was upset by initial finger stick (47%, most of whom were parents of children younger than 5 years), lack of proof that intervention for high cholesterol makes a difference in children (40%), inconvenience (40%), and decision to try a low-fat, low-cholesterol diet before having the child’s lipids rechecked (40%). Family history of hypercholesterolemia, angina, and stroke were not consistently associated with higher follow-up.130
20 Identification of dyslipidemia in children was associated with increased parental reporting (not confirmed with objective measures) of behavior disturbances in the subsequent 12 months, worse scores on the child behavior checklist, but no differences in depression or anxiety for either the children or the mothers.213 Children with dyslipidemia (n=36) and matched controls from one large pediatric practice showed no differences measures of well-being, depression and behavior.215 Older dyslipidemic children (ages 6- 10) had more negative health beliefs than younger children (ages 4-6), and younger children’s reports of parent’s dietary control were negatively related to the children’s feelings of acceptance.212 Parents of children diagnosed with FH reported perceptions of burdensome follow-up causing conflicts between themselves and their children, and difficulty changing diet. Although the older children interviewed as part of this study tended to remember feeling sad, angry, or worried about heart problems, none reported that these worries interfered with their usual activities.214 Children with greater problem solving skills have demonstrated better compliance with recommended diet and greater reductions in LDL-C following traditional diet counseling.216
Limitations of the literature around adverse effects of screening include small sample sizes, mostly observational studies, and the hypothetical nature of risks in the screened populations.
Key Question 4. In Children/adolescents, What is the Effectiveness of Drug, Diet, Exercise, and Combination Therapy in Reducing the Incidence of Adult Dyslipidemia, and Delaying the Onset and Reducing the Incidence of CHD-related Events (Including Optimal Age for Initiation of Treatment)?
No studies meeting inclusion criteria evaluated the effect of a childhood intervention on the incidence of adult dyslipidemia, or the onset or incidence of CHD-related events and outcomes.
Key Questions 5 - 8. What is the Effectiveness of Drug, Diet, Exercise, and Combination Therapy for Treating Dyslipidemia in Children/adolescents?
Summary
Statins and bile acid binding resins are effective for reducing TC and LDL-C in children with FH, the only population for whom they have been studied. These trials might not generalize to children without FH. Trials of other agents are methodologically limited. Diet supplements (psyllium, oat, sterol margarine) and counseling are marginally effective in lowering lipid levels in children with FH or FCH, as well as in children with idiopathic (risk factor associated) dyslipidemia. Trials show minimal if any improvement
21 in lipids with exercise among children from the general population (overweight or normal weight). No trials evaluate exercise treatments in children with FH.
Evidence
Forty randomized controlled trials met inclusion criteria and are included in this report (Evidence Table 4).37-41, 66-69, 71-78, 80-82, 217-236 Of these, ten are rated good quality,37, 38, 40, 69, 71, 76, 78, 218-220 12 fair,41, 66, 67, 72, 74, 81, 221, 222, 225, 232, 233, 235 and 18 poor.39, 68, 73, 75, 77, 80, 82, 217, 223, 224, 226-231, 234, 236 Drug therapy interventions included bile acid binding resins, anti- oxidants (vitamin C and E), docosahexaenoic acid (DHA), HMG-CoA Reductase Inhibitors (statins), and combination therapy. Studies of other types of interventions included exercise therapy and advice, dietary products (margarine, flour, cereal, garlic extract), dietary advice, and combinations of these. Major limitations include 20 or fewer subjects in each arm,39, 67, 73, 82, 221, 223, 224, 230, 234-236 high loss to follow-up,77, 81, 222 lipid measures that are non-standardized or not reported as such, varied inclusion criteria (Table 4B), failure of blinding,77, 220, 228, 231-233 lack of results presented for the period prior to crossover,39, 41, 73-75, 223, 224, 227, 234, 236 and lack of intention to treat analyses.41, 66, 67, 72, 75, 77, 81, 82, 217, 222, 224, 226, 228, 229, 231-233 In addition, one trial did not report data for the placebo group.66
Statins, bile-acid binding resins, and fibrates have been tested only in children with FH and FCH. Applicability of results from these trials to children without these conditions may be difficult. We assessed external validity for each study using the following criteria: population-based (vs recruited from a lipid clinic), presence of a run-in period, number of participants, and reimbursement for participants. Eleven studies were population-based,67, 69, 72, 78, 219-222, 225, 230, 233 while others used populations recruited from single lipid clinics.37, 39-41, 66, 68, 73-77, 80, 82, 223, 224, 226, 231, 237 Fourteen reported the presence of a run-in period,37, 39-41, 66, 67, 69, 71-76, 78, 80, 82, 223, 224, 228, 234-236 one of which did not exclude any participant after the run-in period.66 Five trials failed to report or were unclear about the presence of a run-in period.227, 229-231, 238 Three studies reported reimbursement or free medication for participants;75, 225, 226 others did not mention this. Only 24 studies had 50 or more participants.37, 38l, 40, 66, 68, 69, 71, 72, 76, 78, 80, 81, 217-220, 222, 225- 229, 232, 233 Pharmaceutical or cereal company funding was reported in 14 studies; 16 other studies did not report funding source.
Studies in children with probable or definite familial hypercholesterolemia
Drug treatment. Eighteen trials meeting inclusion criteria evaluated drug therapies for treatment of children with probable or definite heterozygous familial hypercholesterolemia (Table 8).37, 38, 66, 68, 69, 71-76, 78, 80-82, 224, 234, 236 Most of these studies included children who were already compliant with a recommended low-fat, low- cholesterol diet. These diets are presented in Table 1. Both treatment and control groups were maintained on the diet during these trials.
22 All the trials of statin drugs, including one trial of atorvastatin,76 three of lovastatin,38, 72, 78 two of pravastatin37, 71 and three of simvastatin,66, 68, 69 demonstrated improvement in TC and LDL-C among children and adolescents with FH. The decrease in TC ranged from -17% to -32% compared to baseline for treatment groups vs. changes of +3.6% to – 2.3% for placebo. The decreases in LDL-C ranged from –19% to –41% for treatment groups, vs. changes of +0.67% to –3% for placebo. HDL-C decreased in the one trial of atorvastatin,76 increased in one72 but not the other trial of lovastatin,78 did not change in the one trial of pravastatin that reported HDL-C,71 and increased in two trials of simvastatin66, 68 but not the third.69 Overall, HDL-C increased between +1% and +11% for treatment groups, vs. –1% to +4.8% for placebo groups. Triglyceride changes varied, with only simvastatin demonstrating lower triglycerides in all three trials that evaluated it.66, 68, 69 Overall, the range of triglyceride change with treatment was +9% to –24%, vs. placebo +4% to –12%. Results from eight trials of statins, six good,37, 38, 69, 71, 76, 78 one fair,66 and one poor68 quality, were included in a meta-analysis (Figures 7-9). Data from studies reporting triglyceride outcomes were not sufficient for meta-analysis. Pooled estimates for all doses combined included 24.4% (95% C.I. 19.5, 29.2) mean reduction in TC, 30.8% (95% C.I. 24.1, 37.5) mean reduction in LDL-C, and 3.3% (95% C.I. 1.0, 5.7) mean increase in HDL-C. Using drug-adjusted equivalent doses showed linear relationships between dose and drug effect as well as between dose and mean age of the study participants (lower doses associated with younger age, p=0.90, p=0.0029). Other study level covariates, such as duration and percent male or female, were not associated with drug effects. Exclusion of the poor quality study did not affect the results.
One fair quality trial of cholestyramine81 and one poor quality trial of colestipol80 demonstrated decreased total cholesterol and LDL-C but no change in HDL-C or triglycerides. One fair quality study without pre-crossover results demonstrated that patients ages 10-18 preferred and were more compliant with a pill form vs. powder form of cholestyramine.74
Poor quality trials evaluating bezafibrate,82 vitamins C and E,224 DHA,234, 236 p- aminosalicylic acid,73 and a head-to-head comparison of combined colestipol and pravastatin vs. colestipol alone,75 failed to report pre-crossover data. Trials of vitamin C and E and DHA included children and adolescents with FH as well as FCH.
Non-invasive measures of vascular outcomes were evaluated in one good quality study that showed regression of mean carotid intima-media thickness (IMT) over two years with treatment with pravastatin,37 and in two poor quality trials without pre-crossover data that showed improvement in brachial artery flow-mediated dilation with vitamins C and E treatment, and DHA.224, 234
Diet treatment. Of the five trials meeting inclusion criteria that evaluated diet treatments in children with FH (Table 9), one was rated good quality,40 three fair,41, 67, 235 and one poor.39 These trials evaluated plant sterol margarines vs. placebo margarine,39-41, psyllium-enriched cereal,67 and garlic extract.235 Although all three trials of sterol
23 margarines were crossover trials without pre-crossover results presented, the wash-out periods between treatment periods were 4-6 weeks in two of these trials, making it more likely that the results of these trials are valid.40, 41 For these two studies, the mean decreases in TC were -7.4% to –11%, the decreases in LDL-C were –10% to –14%. The differences in HDL-C and triglyceride levels were not significant. The third trial of sterol margarine had no washout period between the treatment arms, but showed similar results (-11% change in TC, -15% for LDL-C, non-significant changes for HDL-C and TG).39 The trial of psyllium-enriched cereal found a 8% decrease in TC in the treatment group vs. –3% in the placebo group, and a decrease in LDL-C of –10% in the treatment group vs. –0.5% in the placebo group, with non-significant post-crossover results for both HDL-C and TG.67 There was no significant improvement in lipid levels with 8 weeks of garlic extract treatment.235
Exercise treatment. No studies meeting inclusion criteria evaluated exercise treatment for lipid lowering in children with FH.
Studies in children with elevated lipids, but not meeting criteria for familial hypercholesterolemia
Drug treatment. No studies meeting inclusion criteria evaluated drug interventions in children without monogenic dyslipidemia.
Diet treatment. Dietary interventions in general populations of children and adolescents were addressed in eight studies, three of good quality,218-220 and five of poor quality (Table 10).77, 223, 227, 229, 231 Two of the good quality studies were from the multi-center Dietary Intervention Study in Children trial. 218, 219 These and other trials77, 227 tested the effect of dietary advice for families and children on dietary outcomes and lipid levels. A good quality trial by the DISC Collaborative Research Group showed that intensive counseling and follow-up over three years reduced TC and LDL-C and improved dietary intake.219 Lipid lowering results were not sustained at five and seven year follow-up, after the intervention ceased, but improvements in diet remained significant.218 One good quality study of the Parent-Child AutoTutorial (PCAT) program220 reported 8% improvement in LDL-C (p<0.05) compared to the at-risk control group. Another poor quality study of PCAT found decreases in both treatment and control groups, but no between-group differences.77 A poor quality study of single vs. multiple nutritional advice sessions found 7-8% decreases in TC and LDL-C for the multiple session group, vs. 3% decrease in TC in the single session group.227
Dietary supplements trials included two poor quality studies of psyllium 223, 229, and one poor quality study of oat bran.231 Psyllium supplements decreased TC and LDL-C (5% and 9% greater decreases in fiber group over control using Step I diet) in the one trial,229 but had no effect in the other,223 The trial of oat bran showed no significant changes in lipids.231
24 Exercise treatment. Six studies, three fair quality225, 232, 233 and the three poor quality226, 228, 230 meeting inclusion criteria evaluated exercise in normal or obese children with elevated lipids (Table 11). The poor studies were rated poor because of differential or low completion rates (51-52%), small numbers of participants (<20) resulting in low generalizability, among other problems (lack of blinding, lack of intention to treat).
In these studies, lipid levels were lower than those typically used to define FH (LDL- C<160mg/dL and TC<240 mg/dL). Two trials comparing supervised, scheduled sessions of aerobic and fitness training to control groups who received no specific regimen showed improvements in HDL-C for the intervention group.230, 233 The others showed minimal or no change in lipids compared to control groups.225, 226, 228, 232
A meta-analysis of these six exercise trials in healthy and/or overweight children and adolescents without identified FH or monogenic dyslipidemia showed non-significant increases in TC, LDL-C and HDL-C (Figures 10-12). This result did not change with exclusion of the poor quality studies, or with exclusion of the study of college-age women (an outlier because of age). However, trials were problematic because baseline lipid levels were different between the treatment and control groups,225, 230, 233 suggesting that randomization may have been flawed. In these studies, the magnitude of difference in the baseline difference between the two groups could be larger or as large as the difference between baseline and endpoint.225, 230, 233 These trials generally had small sample sizes and may not have had enough power to detect a significant effect. Also, most trials only reported the mean lipid level and its associated variance (or S.D.) at baseline and endpoint for both treatment and control groups, and our method to calculate standard error of percent change may have yielded a conservative estimate because the correlation between baseline and endpoints is ignored in the calculation (see Appendix 6).
Combination Diet and Exercise Treatment. Three trials meeting inclusion criteria, two fair221, 222 and one poor,217 evaluated combined regimens of diet and exercise in this population (Table 12). One study showed a 23% increase in HDL-C for the exercise, diet plus behavior change group compared with no significant changes for either the diet plus behavior change or the control groups.221 One “Know Your Body” curriculum study encouraged adoption of regular exercise programs and an AHA prudent diet,239 and found a small decrease in TC that became significant after adjustment for age, gender and race.222 The third trial combined the diet alone and diet plus exercise groups for analysis and found a 4% decrease in TC in the intervention groups compared to a 2.2% increase in the waiting list control group (p=0.03), a 20% increase in HDL-C compared to 6.5% increase in the control group (p=0.007), and a 41% decrease in TG for the intervention group vs. an 8% decrease for the control group (p=0.01).217
25 Key Question 9. What are the Adverse Effects of Drug, Diet, Exercise, and Combination Therapy in Children/adolescents?
Summary
Statin drugs are associated with elevations in liver enzymes and creatinine kinase (CK), and bile-acid binding resins are associated with non-serious gastrointestinal side effects and subclinical decreases in serum vitamins and minerals. There are reports of growth failure in children and adolescents on low-fat diets, particularly those who were placed on low-fat diets without formal nutritional assessment and advice. Most studies of children over two years old on low-fat diets reported normal growth and development. Studies were variable in the quality of adverse effect assessment and reporting, and too short to determine long-term effects.
Evidence
Tables 13-17 and Evidence Table 5 provide details about adverse effects reported in studies of drug, diet, exercise, and combination therapies in children with elevated lipid levels. Data on safety of longer-term and more widespread use of statin drugs in adults are available in the recent Drug Class Review on HMG-Co Reductase Inhibitors, a systematic review of adverse effects of statins conducted by the Oregon EPC and updated in June 2005.50
Drug treatment
Information about adverse events was reported in 15 studies of statins (Table 13),37, 38, 66, 68-72, 76, 78, 240-244 in 22 studies of bile-acid binding resins (Table 14) reported in 23 publications,74, 75, 80, 81, 245-263 and in 9 studies of various other drugs or drug combinations (Table 15).54, 73, 82, 264-269 Studies used randomized controlled trial, open-label trial, non- comparative, retrospective, and other descriptive designs. Where possible, prevalence rates are reported for studies with more than 20 participants.
Statins were associated with increased ALT and AST in three studies (0-3% of participants in RCTs),69, 76, 241 while six other studies found no changes in liver enzymes overall.37, 68, 78, 240, 242, 243 CK levels were elevated in five studies (range 0.5-4% reporting a significant change in CK, with one study reporting 45% having any abnormal CK test, but none statistically significant),69, 71, 72, 241-243 were normal in four studies,37, 68, 78, 240 and subjects reported myalgia in four studies.69, 240-242 Normal sexual development was reported in four studies.37, 38, 76, 78 No studies observed an adverse effect of statins on sexual development or endocrine function, although one study observed an 18% increase in dehydroepiandrosterone sulfate (DHEA) levels with lovastatin compared with a 5% increase with placebo.78 Other adverse effects of statins include gastrointestinal complaints in four studies,69, 71, 240, 241 rash/pruritus in two studies,69, 71 and headache in
26 three studies.69, 71, 241 A study of psychological effects found no adverse effects associated with statin therapy.70 Statins have known drug interactions with erythromycin which can precipitate rhabdomyolosis.50, 76
Bile-acid binding resins were associated with gastrointestinal complaints (8-26%), such as flatulence and constipation, in 11 studies,74, 75, 80, 81, 247, 250, 252, 254, 256, 259, 260 and their unpalatability (up to 50%) was noted in seven studies.248, 252, 253, 255, 258-260 Reduced serum folate occurred with colestipol in one study 80 and with cholestyramine in five studies,81, 247, 256-258 one of which reported reduced folate occurring only in the females.247 Other studies reported no reduction of serum folate after 2 to 3 years of treatment with colestipol.250, 253 One study of cholestyramine257 and one study of colestipol253 reported a significant decrease in serum levels of vitamins A and E and inorganic phosphorus, but others demonstrated normal vitamin A and E levels over time.260, 262 One study of cholestyramine reported transient increases in LDH, and abnormalities in SGOT that persisted for 6 month,247 but others showed normal liver function tests.260, 262 One study reported a child whose height for age dropped below –2 S.D. while on colestipol (1 S.D. = 2.4 cm).249 Growth was reported normal in seven studies,80, 81, 250, 251, 256, 257, 261 including sexual maturation over 4.3 years of treatment.261 There was one case report of extensive loss of dental enamel in a 7-year-old boy after two years of cholestyramine use. This is thought to have resulted from mixing cholestyramine in Kool-Aid and the child holding the mixture in the mouth for 10 to 15 minutes before swallowing.245 Serum calcium, phosphorus folate, and vitamin B12 values, and results of a radioactive bone mineral analysis were normal.
Other drugs studied were fenofibrate, bezafibrate, niacin, dextrothyroxine (D-T4), orlistat, P-aminosalicylic acid, and probucol. These studies enrolled small numbers (7- 30, mean 15.8) of subjects. A study of fenofibrate reported significant decreases in total alkaline phosphatase, uric acid, inorganic phosphates, and bilirubin, and increases in ALT and AST levels.269 A study of bezafibrate observed high alkaline phosphatase in one child.82 The combination of bezafibrate and sitosterol resulted in decreased alkaline phosphatase and increased transferrin levels (by 20%) in two subjects.264
Two of two studies of niacin reported increased liver enzymes (6 of 21 children in one study), and multiple other symptoms (flushing, abdominal pain, nausea, headache).266, 268 One seven year-old girl developed an influenza-like febrile illness with serum aminotransferase levels >400 IU/L two days after starting niacin. Niacin was considered a possible cause and was discontinued.266 In another study of niacin combined with a bile-acid binding resin, one participant developed clinical symptoms of hepatotoxicity.268
P-aminosalicylic acid was associated with mild gastric irritation.73 A study of probucol reported nausea in one of seven patients, but probucol was otherwise well-tolerated, and growth and development were normal after 15 to 21 months of treatment.54 Orlistat was associated with a significant drop in 25-hydroxy-vitamin D levels after one month.267 Three of 20 subjects enrolled in this study required additional vitamin D supplementation despite the prescription of a daily multivitamin containing vitamin D. In a study of the effects of D T4 on the pituitary-thyroid axis of hypercholesterolemic children, an increase
27 in the basal level of T3 was observed after treatment with D-T4, and the expected stimulation of TSH and T3 secretion in response to TRH was absent.265
Low-fat diet
Nineteen studies of dietary fat restriction reported effects on growth, nutrient intake, laboratory safety parameters, or other adverse effects (Table 16).218, 219, 227, 270-285
Twelve studies reported normal height growth,218, 227, 271-273, 276, 277, 279, 280, 282-284 although weight loss occurred among three children in two of these studies.272, 279 Weight loss in one subject was caused by anorexia nervosa, and in another by a parent-instituted vegetarian diet that was more restrictive than recommended.279
Growth failure in one study occurred among 8 of 40 (20%) children with dyslipidemia, 3 (7.5%) of whom had nutritional dwarfing and no progression of puberty.278 Those with growth failure were ages 7-13 at diagnosis. In this study, families received general advice to reduce dietary fat and cholesterol at the time of diagnosis, but the specific recommendations varied in each case and implementation by the families was unsupervised. The period of unmonitored dietary restriction ranged from 2 weeks to 4.5 years and those with nutritional dwarfing had longer periods of time between diagnosis and formal dietary assessment and counseling.278 Total energy and zinc consumption were lower in those with growth failure. All three children with nutritional dwarfing had inadequate intake of multiple different vitamins and minerals. Other studies reported inadequate intake of vitamin D,274 vitamin E,219, 285, and zinc.219 Failure to thrive has been demonstrated in children under age two eating fat-restricted diets.286 Fat-restricted diets are not recommended for children under age two because fat and cholesterol are necessary for normal growth and development in children under age two.35
A study of lymphocyte T subset counts found significant decreases in CD3, CD4, and CD8 after six months of dietary therapy.275 The changes in CD3 and CD8 counts were significantly correlated with changes in triglyceride serum levels. In all cases, lymphocyte T subset counts remained within normal ranges. The other immune indexes (immunoglobulins G, A, M, and complements C3, C4, and Factor B) did not change significantly.
Two studies of the psychological effects of diet therapy reported no adverse effects in academic function, psychological symptoms, and family function,281 or in behavioral and emotional assessment scores.271
A retrospective study of familial chylomicronemia observed patients in whom dietary intervention since infancy had provided at most 10% of energy as long-chain triglycerides.282 The study observed abnormal values for serum iron, alkaline phosphatase, and total calcium. Two children experienced severe abdominal pain, and an adolescent female developed acute pancreatitis after using an oral contraceptive agent.
28 Dietary supplements
Fourteen studies provided information about the adverse effects of various dietary supplements, including fiber, rapeseed oil, plant sterol or stanol esters, fish oil, soy- protein beverage, garlic extract, locust bean gum, and antioxidant vitamins (Table 17).39, 41, 223, 235, 287-296
Two children (4% of the treatment group) reported abdominal discomfort using fiber tablets that contained 50% wheat bran and 50% pectin, administered at 100-150 mg/kg/day.223, 290, 293 There were no adverse effects with psyllium fiber in two other studies.223, 290
Studies of rapeseed oil alone,291 or in combination with sitostanol39 did not report adverse effects, and both studies reported that the diets were well-tolerated with good compliance. A plant sterol ester spread that provided 1.6 g/day of sterol ester reported an increase in ALT of 16.8% of participants.41 A longer study of plant sterol ester spread reported increased retinol levels, but no increase in transaminases.287 Sitosterol and sitostanol given in succession decreased alkaline phosphatase.288 In a study of fish oil (5 g/day), epistaxis occurred in eight of 11 subjects, thought to be secondary to decreased platelets that can occur with omega-3 fatty acids. Prothrombin and partial thromboplastin times and platelet counts were normal, and there was no evidence of liver dysfunction.289 After adjustment for lipid changes, lycopene was 8.1% lower (p=0.015), and serum levels of retinol were 15.6% higher (p<0.001), compared with controls. Studies of antioxidant vitamin therapy,292 substitution of a soy-protein beverage for cow’s milk,294 and garlic extract235 showed no significant adverse effects. A study of locust bean gum food products reported increased rectal gas that resolved after 1 to 2 weeks, with no other adverse effects or abnormal lab values.296
Exercise
A school-based program examined the effect of supervised exercise training on the lipid profiles of normal prepubertal children and reported 100% adherence and no adverse effects.297 In another study, treadmill tests elicited an exaggerated blood pressure response in boys with dyslipidemia.298 These subjects had significantly higher blood pressures after exercise compared with normal controls (systolic 182 vs. 160 mmHg, p<0.0003; diastolic 77 vs. 9 mmHg, p<0.03), and their systolic blood pressures remained significantly higher at the end of recovery (120 vs. 112 mmHg, p<0.005).
Key Question 10. Does Improving Dyslipidemia in Childhood Reduce the Risk of Dyslipidemia in Adulthood?
No studies were identified that directly evaluated whether treatment of idiopathic dyslipidemia in childhood reduces the risk of dyslipidemia in adulthood. Indirect
29 evidence includes: Tracking of lipid during childhood and adolescence (40-55% remain in the highest quintile); and data from young adults whose CHD risk 22-42 years later was associated with their baseline lipid levels.4 Identification and treatment of previously undiagnosed secondary dyslipidemia (e.g. hypothyroidism) can improve lipids. Children with FH continue to fully express the phenotype in adulthood and the natural history of this disease is early CHD. Thus, in FH populations, long-term placebo controlled trials of drug therapy may be unethical.
Key Question 11. What are the Cost Issues Involved in Screening for Dyslipidemia in Children/adolescents?
Seven studies addressed the cost of screening children and adolescents. These studies reported overall costs of screening but did not relate it to effectiveness of the screening test, or take a societal perspective on the analysis of cost.130, 177, 178, 182, 183, 299, 300 The only cost-effectiveness study identified was based in England and used a simulated population ages 16-54.301 This study determined that population screening of 16 year olds only was as cost-effective as family tracing. However, these results may not be applicable to the U.S. No cost-effectiveness studies addressed screening in children younger than 16.
30 IV. DISCUSSION
Conclusions
A summary of the evidence included in this review is presented in Table 18. Overall, there was inadequate evidence to address key questions about efficacy of screening children and adolescents for dyslipidemia for delaying the onset and reducing the incidence of CHD-related events, effectiveness of treatments (drug, diet, exercise and combination) on reducing incidence of adult dyslipidemia or delaying the onset and reducing the risk of CHD-related events, whether improving dyslipidemia in children and adolescents reduces the risk of adult dyslipidemia, and cost-effectiveness of screening for dyslipidemia in children and adolescents.
Normal values for lipids in children are currently defined according to population levels (percentiles). The NCEP has defined levels of LDL-C for which drug treatment (LDL-C >190 mg/dL or LDL-C >160 mg/dL with family history of early CHD), further evaluation, diet therapy and testing (LDL-C >130 mg/dL), and diet therapy with increased surveillance (LDL-C 110-129 mg/dL) are recommended. There is concern that the LRC values that have defined normal lipid levels in children were performed with such rigor (assured fasting, research coordinators to facilitate follow-up) that they may not be applicable to a general screening population where assessments are less controlled. Also, demographic shifts toward a more overweight population of children may make the LRC values less representative of the current population.
The most appropriate screening test is one that accurately predicts future risk and benefit from treatment. Screening recommendations for children with a family history of parental high cholesterol include an initial TC followed by a repeat TC and/or lipoprotein profile depending on the degree of elevation in TC. In practice, screening is often performed using TC and HDL-C, rather than TC alone follow-up by lipoprotein analysis if indicated by guidelines. Using LRC data, fasting TC > the 95th percentile detected an LDL-C above the 95th percentile for age and gender with 69% sensitivity and 98% specificity. However, recent data suggest that no single value of TC is sufficient to place a child confidently in the NCEP risk categories. Assessments of the diagnostic accuracy with which LDL-C levels predict familial hypercholesterolemia (FH) have been performed in known FH families, but not tested in the general population of children. The gold standard for diagnosis of FH is mutation analysis or fibroblast LDL-C receptor activity level testing. These confirmatory tests have not been applied to population screening trials.
Studies of lipid level tracking within childhood and from childhood to adulthood demonstrate large ranges of correlation. Serial correlations measured in individual children over time are more consistently higher for TC (r=0.38-0.78) and LDL-C (r=0.4- 0.7), than for HDL-C (r=0-0.8) and TG (r=0.1-0.58). Approximately 40-55% of children with elevated lipids (by percentile) will continue to have elevated lipids on follow-up. Of
31 those in the highest quintile of TC at follow-up, 15-19% had TC levels in the 1st or 2nd (lowest) quintiles at baseline.
Evidence for risk factors other than family history for predicting dyslipidemia in children is strongest for overweight, but otherwise lacking. Epidemiologic cross-sectional and cohort studies establish a strong statistical association between overweight and dyslipidemia. A BMI > 95th percentile, has been shown to be a stronger correlate of cumulative CHD risk (OR=19.2; 95% CI 7.6-48.5) than hyperinsulinemia. However, the magnitude of the effect of overweight on lipids, and the potential impact of incorporating overweight into a screening tool for dyslipidemia have not been addressed. Some risk factors may be useful for identifying groups of children and/or adolescents with idiopathic dyslipidemia. Multiple other risk factors (diet, physical inactivity, and aerobic capacity/fitness) have not been evaluated adequately to assess their contribution to dyslipidemia in children or their usefulness as screening tools either alone or in combination.
Currently recommended screening strategies have low adherence by providers, and limited compliance by parents and children. No trials compare strategies (location, venue, age, provider) for screening in children and adolescents. No studies address the frequency and optimal age for testing. Family history questions have limited diagnostic accuracy and have not been universally agreed upon. Studies of screening in children and adolescent and NCEP evaluate measures of TC, while adult guidelines focus on TC and HDL-C. Some experts suggest applying TC and HDL-C screening tests to children and adolescents.
There is a growing literature on the association between lipid levels in children and non- invasive vascular outcomes. Carotid intima-media thickness (IMT) is significantly higher among overweight children compared to non-overweight children,302 and among children with FH compared with non-affected siblings.83 Impaired nitrate dilation is lower and brachial IMT higher in children with diabetes and FH compared with healthy children.303 Data from the Muscatine cohort demonstrate that measurement of adult carotid IMT correlates with lipid measurements taken in childhood, and those with TC in the upper quartile as measured at age 8-18 had an OR of 1.53 (men) and 1.43 (women) for a 1 standard deviation increase in carotid IMT.304 Further exploration into arterial IMT as a risk factor identifiable in children may be warranted to evaluate its usefulness in screening and/or decisions about treatment.305
Adverse effects of screening include parental non-compliance with screening and follow- up recommendations. This is reported to be related to concerns about test accuracy, lack of proof that intervention makes a difference in children, concerns about upsetting the child, refusal by the child, inconvenience, or decisions to institute a diet themselves and have the child rechecked subsequently.
Drug treatment for dyslipidemia in children has been studied only in children with FH, the population for whom these drugs are FDA-approved and recommended by the NCEP. Statins are effective for reducing TC and LDL-C in children with FH. It is not clear how
32 this efficacy translates to children with milder and/or non-monogenic dyslipidemia (i.e. idiopathic dyslipidemia), and it is not known how frequently these medications are used in children without FH in practice. Studies of bile acid binding resins demonstrate improved TC and LDL-C, but were few and fair/poor quality. Niacin has been used, but is not FDA approved for children.266, 306 Target lipid levels for children who warrant treatment for dyslipidemia are outlined in the NCEP and AAP guidelines (Fig 1); the recently updated adult (ATP III) guidelines suggest more aggressive targets for adults that are not applicable to children.
Diet supplements (psyllium, oat, sterol or stanol margarines) and advice were minimally effective in both children and adolescents with FH or FCH and those without. Exercise treatments provided little to no improvement in children and adolescents without identified monogenic dyslipidemia. There were no RCTs of exercise in FH populations.
Adverse effects of treatment are reported in controlled and non-controlled studies of drug, diet, exercise, and combination therapy in children and adolescents. Statin drugs were associated primarily with elevations in LFTs and CK. Bile-acid binding resins were associated with GI side effects and decreased levels of serum vitamins and minerals. Niacin has not been systematically studied in children. Low-fat diets have been associated with growth retardation, nutritional dwarfing, and inadequate calorie and nutrient intake in children placed on a low-fat diet prior to formal nutrition counseling.278, 285 Other studies report normal growth and development in children over 2 years of age on monitored low-fat diets. Few side effects other than elevated blood pressure in children with FH were noted with exercise. The duration of follow-up in these studies ranged from 10 days to eight years. Studies were generally not of sufficient duration to determine long-term effects of either short or extended use.
Limitations of the Literature
In assessing the evidence for screening and treatment of children with dyslipidemia, there remain several areas of uncertainty. Current screening guidelines are based on population norms obtained 30 years ago and do not take into account age, sex, race differences and temporal trends. They are based on family history assessments that have been shown to miss large numbers of children with dyslipidemia and that might lead to bias against detecting elevated lipids in African American children and adolescents, or children and adolescents from single-parent families. Tracking studies follow large percentiles (quartiles, quintiles) rather than focusing on those children and adolescents above the 95th percentile who are likely to have the highest risk. Risk factors that might contribute to a risk assessment tool have not been adequately studied. Screening strategies including methods for assuring adequate assessment of minorities and those with unknown family history, and improving provider and parental compliance with guidelines deserve attention. Studies of statins include only children with FH, are relatively small and of short duration. There are no trials with long-term follow-up for adult lipid outcomes or CHD-related events. Studies of treatments (drug, diet and exercise) for dyslipidemia
33 could be improved by use of more rigorous study designs, enrollment of larger population-based samples and systematic reporting of adverse effects.
In summary, although many studies about various aspects of dyslipidemia in children have been published, there are conclusive answers for few of the questions addressing primary care providers when faced with screening the general population of children and adolescents. Good evidence exists for the efficacy of statin treatments in children and adolescents with FH and the unreliable diagnostic accuracy of family history for detecting high lipid levels in children. Although current literature demonstrates imperfect tracking of TC and LDL-C through childhood and from childhood to adulthood, there are no studies evaluating tracking for those with lipids above the 95th percentile. Guidelines remain based on LRC data rather than updated population norms.
Future Research
Directions for future research should include identification of the impact that risk factors other than family history (such as overweight, physical inactivity) have on lipids, in order to develop risk assessment tools. Such tools may provide a better indication of a child’s actual risk, and could serve to facilitate screening by narrowing the number of children requiring serum lipid testing. Further research into non-invasive screening strategies for screening, such as arterial IMT may be useful.304 Randomized controlled clinical trials of screening strategies to determine those that may be more effective than those currently in practice, both in terms of parental compliance and provider adherence to guidelines are important. There should be continued follow-up of currently established cohorts to assess the impact of screening for dyslipidemia in childhood on adult CHD outcomes.
To further assess the efficacy and safety of treatment options in children and adolescents with dyslipidemia, long-term follow-up of children treated with statins for impact of sustained improvement of lipid in childhood on adult lipid levels, adult CHD-outcomes and long-term safety will be useful. The impact of exercise on lipid levels should be evaluated further, particularly in children with lipid levels above the 95th percentile. There is a need for improved reporting of adverse effects in trials and use of standardized methods for reporting so that data can be combined across trials. Cost-effectiveness of universal screening vs. screening via family history as currently recommended should be evaluated.
34 REFERENCES
1. Ahmed SM, Clasen ME, Donnelly JE. Management of dyslipidemia in adults. Am Fam Physician. 1998;57(9):2192-2204. 2. Dorlands Illustrated Medical Dictionary. 29th ed. Philadelphia, PA: W.B. Saunders Company; 2000. 3. Kelder SH, Osganian SK, Feldman HA, et al. Tracking of physical and physiological risk variables among ethnic subgroups from third to eighth grade: the Child and Adolescent Trial for Cardiovascular Health cohort study. Prev Med. 2002;34(3):324-333. 4. Klag MJ, Ford DE, Mead LA, et al. Serum cholesterol in young men and subsequent cardiovascular disease. N Engl J Med. 1993;328(5):313-318. 5. Eyre H, Kahn R, Robertson RM, et al. Preventing cancer, cardiovascular disease, and diabetes: a common agenda for the American Cancer Society, the American Diabetes Association, and the American Heart Association. Circulation. 2004;109(25):3244-3255. 6. Morrison JA. A longitudinal evaluation of the NCEP-Peds guidelines for elevated total and LDL cholesterol in adolescent girls and boys. Prog Pediatr Cardiol. 2003;17:159-168. 7. Whitlock EP, Williams SB, Gold R, Smith P, Shipman S. Screening and interventions for childhood overweight: systematic evidence synthesis. Evidence Synthesis #36. Rockville, MD: Agency for Healthcare, Research, and Quality; 2005. 8. Srinivasan SR, Myers L, Berenson GS. Predictability of childhood adiposity and insulin for developing insulin resistance syndrome (syndrome X) in young adulthood. Diabetes. 2002;51:204-209. 9. Ford ES, Ajani UA, Mokdad AH. The metabolic syndrome and concentrations of c-reactive protein among U. S. youth. Diabetes Care. 2005;28(4):878-881. 10. Goodman E, Dolan LM, Morrison JA, Daniels SR. Factor analysis of clustered cardiovascular risks in adolescence: obesity is the predominant correlate of risk among youth. Circulation. 2005;111:1970-1977. 11. Kavey R-EW, Daniels SR, Lauer RM, Atkins DL, Hayman LL, Taubert K. American heart association guidelines for primary prevention of atherosclerotic cardiovascular disease beginning in childhood. J Pediatr. 2003;142(4):368-372. 12. Williams CL, Hayman LL, Daniels SR, et al. Cardiovascular health in childhood: A statement for health professionals from the Committee on Atherosclerosis, Hypertension, and Obesity in the Young (AHOY) of the Council on Cardiovascular Disease in the Young, American Heart Association. Circulation. 2002;106(1):143-160. 13. Oregon Evidence-based Practice Center (EPC). Screening for Dyslipidemia in Adults: Brief Update of 2001 USPSTF Review. Rockville, MD: Agency for Healthcare, Research, and Quality; 2006 (in press). 14. The Lipid Research Clinics Population Studies Data Book. Vol 1. Bethesda, MD: National Institutes of Health; Lipid Metabolism Branch; Division of Heart and Vascular Diseases; National Heart, Lung, and Blood Institute.; 1980. 15. Srinivasan SR, Sharma C, Foster TA, Berenson GS. Longitudinal changes of serum lipid and apoB levels in a newborn-infant cohort. Metab: Clin & Exp. Feb 1982;31(2):167-171. 16. Schrott HG, Bucher KA, Clarke WR, Lauer RM. The Muscatine hyperlipidemia family study program. Prog Clin Biol Res. 1979;32:619-646. 17. Freedman DS, Shear CL, Srinivasan SR, Webber LS, Berenson GS. Tracking of serum lipids and lipoproteins in children over an 8-year period: the Bogalusa Heart Study. Prev Med.1985;14(2):203-216. 18. Freedman DS, Dietz WH, Srinivasan SR, Berenson GS. The relation of overweight to cardiovascular risk factors among children and adolescents: the Bogalusa Heart Study. Pediatrics. 1999;103(6 Pt 1):1175-1182. 19. Bao W, Srinivasan SR, Valdez R, Greenlund KJ, Wattigney WA, Berenson GS. Longitudinal changes in cardiovascular risk from childhood to young adulthood in offspring of parents with coronary artery disease: the Bogalusa Heart Study. JAMA. 1997;278(21):1749-1754. 20. Newman WPI, Freedman DS, Voors AW, Gard PD, et al. Relation of serum lipoprotein levels and systolic bolld pressure to early atherosclerosis. N Engl J Med. 1986;315:138-144.
35 21. Freedman DS, Newman WP, 3rd, Tracy RE, et al. Black-white differences in aortic fatty streaks in adolescence and early adulthood: the Bogalusa Heart Study. Circulation. 1988;77(4):856-864. 22. Berenson GS, Wattigney WA, Tracy RE, et al. Atherosclerosis of the aorta and coronary arteries and cardiovascular risk factors in persons aged 6 to 30 years and studied at necropsy (The Bogalusa Heart Study). Am J Cardiol. 1992;70(9):851-858. 23. Berenson GS, Srinivasan SR, Bao W, Newman WP, 3rd, Tracy RE, Wattigney WA. Association between multiple cardiovascular risk factors and atherosclerosis in children and young adults. The Bogalusa Heart Study. N Engl J Med. 1998;338(23):1650-1656. 24. Strong JP, Zieske AW, Malcom GT. Lipoproteins and atherosclerosis in children: an early marriage? Nutr Metab Cardiovasc Dis. 2001;11(Suppl 5):16-22. 25. Kwiterovich PO, Jr. Prevention of coronary disease starting in childhood: what risk factors should be identified and treated? Coron Artery Dis. 1993;4(7):611-630. 26. Pathological Determinants of Atherosclerosis in Youth (PDAY) Research Group. Relationship of atherosclerosis [in young men] to serum lipoprotein cholesterol concentrations and smoking. JAMA. 1990;264:3018-3024. 27. Goldstein JL, Hobbs HH, Brown MS. Familial hypercholesterolemia. In: Seriver CR, Beaudet AL, Sly WS, Valle D, eds. The metabolic basis of inherited disease. Vol 120. New York: McGraw- Hill; 2001. 28. Clauss SB, Kwiterovich P. Genetic disorders of lipoprotein transport in children. Prog Pediatr Cardiol. 2003;17:123-133. 29. Laskarzewski PM, Khoury P, Kelly K, Mellies MJ, Morrison JA, Glueck CJ. Prevalence of familial hyper- and hypolipoproteinemias in Blacks: the Princeton School District Family Study. Prev Med. 1982;11(2):142-161. 30. Austin MA, Hutter CM, Zimmern RL, Humphries SE. Genetic causes of monogenic heterozygous familial hypercholesterolemia: a HuGE prevalence review. Am J Epidemiol. 2004;160(5):407-420. 31. Kwiterovich PO, Jr. Diagnosis and management of familial dyslipoproteinemia in children and adolescents. Pediatr Clin North Am. 1990;37(6):1489-1523. 32. Cortner JA, Coates PM, Gallagher PR. Prevalence and expression of familial combined hyperlipidemia in childhood. J Pediatr. 1990;116(4):514-519. 33. Cortner JA, Coates PM, Gallagher PR. Prevalence and expression of familial combined hyperlipidemia in childhood. J Pediatr. 1990;116(4):514-519. 34. National Cholesterol Education Program (NCEP). Report of the expert panel on blood cholesterol levels in children and adolescents. Pediatrics. 1992;89(3 Pt 2):525-584. 35. American Academy of Pediatrics. Committee on Nutrition. Cholesterol in childhood. Pediatrics. 1998;101(1 Pt 1):141-147. 36. American Heart Association. Step I, Step II, and TLC Diets. http://www.americanheart.org/presenter.jhtml?identifier=4764. Accessed April 29, 2005. 37. Wiegman A, Hutten BA, de Groot E, et al. Efficacy and safety of statin therapy in children with familial hypercholesterolemia: a randomized controlled trial. JAMA. 2004;292(3):331-337. 38. Clauss SB, Holmes KW, Hopkins P, et al. Efficacy and safety of lovastatin therapy in adolescent girls with heterozygous familial hypercholesterolemia. Pediatrics. 2005;116(3):682-688. 39. Gylling H, Siimes MA, Miettinen TA. Sitostanol ester margarine in dietary treatment of children with familial hypercholesterolemia. J Lipid Res. 1995;36(8):1807-1812. 40. de Jongh S, Vissers MN, Rol P, Bakker HD, Kastelein JJ, Stroes ES. Plant sterols lower LDL cholesterol without improving endothelial function in prepubertal children with familial hypercholesterolaemia. J Inherit Metab Dis. 2003;26(4):343-351. 41. Amundsen AL, Ose L, Nenseter MS, Ntanios FY. Plant sterol ester-enriched spread lowers plasma total and LDL cholesterol in children with familial hypercholesterolemia. Am J Clin Nutr. 2002;76(2):338-344. 42. Tammi A, Ronnemaa T, Miettinen TA, et al. Effects of gender, apolipoprotein E phenotype and cholesterol-lowering by plant stanol esters in children: the STRIP study. Special Turku Coronary Risk Factor Intervention Project. Acta Paediatr. 2002;91(11):1155-1162. 43. U.S. Preventive Services Task Force. Screening for lipid disorders: recommendations and rationale. http://www.ahrq.gov/clinic/ajpmsuppl/lipidrr.pdf. Accessed April 29, 2005. 44. Pignone MP, Phillips CJ, Lannon CM, et al. Screening for lipid disorders. Rockville, MD: Agency for Healthcare, Research, and Quality; 2001. Systematic Evidence Review Number 4.
36 45. National Cholesterol Education Program Expert Panel on Detection E, Treatment of High Blood Cholesterol in A. Third Report of the National Cholesterol Education Program (NCEP) Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III) final report. Circulation. 2002;106(25):3143-3421. 46. National Cholesterol Education Project. Highlights of the report of the Expert Panel on Blood Cholesterol Levels in Children and Adolescents. Pediatrics. 1992;89(3):495-501. 47. Slavin RE. Best practice synthesis: An alternative to meta-analytic and traditional reviews. Education Res. 1986;15:5-11. 48. Harris RP, Helfand M, Woolf SH, et al. Current methods of the US Preventive Services Task Force. A review of the process. Am J Prev Med. 2001;30(3S):21-35. 49. Thompson D, Edelsberg J, Colditz GA, Bird AP, Oster G. Lifetime health and economic consequences of obesity. Arch Intern Med. 1999;159(18):2177-2183. 50. Helfand M, Carson S, Kelley C. Drug Class Review on Statins. Final Report. http://www.ohsu.edu/drugeffectiveness/reports/documents/Statins. Accessed June 15, 2005. 51. National Heart Lung and Blood Institute. Cardiovascular profile of 15,000 children of school age in three communities 1971-1975. Bethesda, MD: U.S. Department of Health, Education, and Welfare, Public Health Service, National Institutes of Health; 1978. 52. Berenson GS. Cardiovascular risk factors in children. New York: Oxford University Press; 1980. 53. National Center for Health Statistics. Serum cholesterol levels of persons 4-74 years of age by socioeconomic characteristics: United States, 1971-74. Hyattsville, MD: U.S. Department of Health Education and Welfare, Public Health Service, National Center for Health Statistics; 1980. 54. Baker SG, Joffe BI, Mendelsohn D, Seftel HC. Treatment of homozygous familial hypercholesterolaemia with probucol. S Afr Med J. 1982;62(1):7-11. 55. Resnicow K, Morley-Kotchen J, Wynder E. Plasma cholesterol levels of 6585 children in the United States: results of the know your body screening in five states. Pediatrics. 1989;84(6):969- 976. 56. Hickman TB, Briefel RR, Carrol MD, et al. Distributions and trends of serum lipid levels among United States children and adolescents ages 4-19 years: data from the third National Health and Nutrition Examination Survey. Prev Med. 1998;27:879-890. 57. Labarthe DR, Dai S, Fulton JE. Cholesterol screening in children: insights from Project HeartBeat! and NHANES III. Prog Pediatr Cardiol. 2003;17:169-178. 58. Kwiterovich PO, Jr., Heiss G, Johnson N, Chase GA, Tamir I, Rifkind B. Assessment of plasma total cholesterol as a test to detect elevated low density (beta) lipoprotein cholesterol levels (type IIa hyperlipoproteinemia) in young subjects from a population-based sample. Am J Epidemiol. 1982;115(2):192-204. 59. Shea S, Basch CE, Irigoyen M, et al. Failure of family history to predict high blood cholesterol among Hispanic preschool children. Prev Med. 1990;19(4):443-455. 60. Dennison BA, Kikuchi DA, Srinivasan SR, Webber LS, Berenson GS. Serum total cholesterol screening for the detection of elevated low-density lipoprotein in children and adolescents: the Bogalusa Heart Study. Pediatrics. 1990;85(4):472-479. 61. Neufeld E. Personal communication to Dr. Elizabeth Haney about screening tests for children. Portland, OR; 2005. 62. Rifai N, Neufeld E, Ahlstrom P, Rimm E, D'Angelo L, Hicks JM. Failure of current guidelines for cholesterol screening in urban African-American adolescents. Pediatrics. 1996;98(3 Pt 1):383- 388. 63. Gillman MW, Cupples LA, Moore LL, Ellison RC. Impact of within-person variability on identifying children with hypercholesterolemia: Framingham Children's Study. J Pediatr. 1992;121(3):342-347. 64. Harris N, Neufeld EJ, Newburger JW, et al. Analytical performance and clinical utility of a direct LDL-cholesterol assay in a hyperlipidemic pediatric population. Clin Chem. 1996;42(8 Pt 1):1182- 1188. 65. Ticho BS, Neufeld EJ, Newburger JW, Harris N, Baker A, Rifai N. Utility of direct measurement of low-density lipoprotein cholesterol in dyslipidemic pediatric patients. Arch Pediatr Adol Med. 1998;152(8):787-791. 66. Couture P, Brun LD, Szots F, et al. Association of specific LDL receptor gene mutations with differential plasma lipoprotein response to simvastatin in young French Canadians with
37 heterozygous familial hypercholesterolemia. Arterioscler Thromb Vasc Biol. 1998;18(6):1007- 1012. 67. Davidson MH, Dugan LD, Burns JH, Sugimoto D, Story K, Drennan K. A psyllium-enriched cereal for the treatment of hypercholesterolemia in children: a controlled, double-blind, crossover study. Am J Clin Nutr. 1996;63(1):96-102. 68. de Jongh S, Lilien MR, op't Roodt J, Stroes ES, Bakker HD, Kastelein JJ. Early statin therapy restores endothelial function in children with familial hypercholesterolemia. J Am Coll Cardiol. 2002;40(12):2117-2121. 69. de Jongh S, Ose L, Szamosi T, et al. Efficacy and safety of statin therapy in children with familial hypercholesterolemia: a randomized, double-blind, placebo-controlled trial with simvastatin. Circulation. 2002;106(17):2231-2237. 70. de Jongh S, Kerckhoffs MC, Grootenhuis MA, Bakker HD, Heymans HS, Last BF. Quality of life, anxiety and concerns among statin-treated children with familial hypercholesterolaemia and their parents. Acta Paediatr. 2003;92(9):1096-1101. 71. Knipscheer HC, Boelen CC, Kastelein JJ, et al. Short-term efficacy and safety of pravastatin in 72 children with familial hypercholesterolemia. Pediatr Res. 1996;39(5):867-871. 72. Lambert M, Lupien PJ, Gagne C, et al. Treatment of familial hypercholesterolemia in children and adolescents: effect of lovastatin. Canadian Lovastatin in Children Study Group. Pediatrics. 1996;97(5):619-628. 73. Malloy MJ, Kane JP, Rowe JS. Familial hypercholesterolemia in children: treatment with p- aminosalicylic acid. Pediatrics. 1978;61(3):365-372. 74. McCrindle BW, O'Neill MB, Cullen-Dean G, Helden E. Acceptability and compliance with two forms of cholestyramine in the treatment of hypercholesterolemia in children: a randomized, crossover trial. J Pediatr. 1997;130(2):266-273. 75. McCrindle BW, Helden E, Cullen-Dean G, Conner WT. A randomized crossover trial of combination pharmacologic therapy in children with familial hyperlipidemia. Pediatr Res. 2002;51(6):715-721. 76. McCrindle BW, Ose L, Marais AD. Efficacy and safety of atorvastatin in children and adolescents with familial hypercholesterolemia or severe hyperlipidemia: a multicenter, randomized, placebo- controlled trial. J Pediatr. 2003;143(1):74-80. 77. Stallings VA, Cortner JA, Shannon BM, et al. Preliminary report of a home-based education program for dietary treatment of hypercholesterolemia in children. Am J Health Promot. 1993;8(2):106-108. 78. Stein EA, Illingworth DR, Kwiterovich PO, Jr., et al. Efficacy and safety of lovastatin in adolescent males with heterozygous familial hypercholesterolemia: a randomized controlled trial. JAMA. 1999;281(2):137-144. 79. Sveger T, Flodmark CE, Nordborg K, Nilsson-Ehle P, Borgfors N. Hereditary dyslipidemias and combined risk factors in children with a family history of premature coronary artery disease. Arch Dis Child. 2000;82(4):292-296. 80. Tonstad S, Sivertsen M, Aksnes L, Ose L. Low dose colestipol in adolescents with familial hypercholesterolaemia. Arch Dis Child. 1996;74(2):157-160. 81. Tonstad S, Knudtzon J, Sivertsen M, Refsum H, Ose L. Efficacy and safety of cholestyramine therapy in peripubertal and prepubertal children with familial hypercholesterolemia. J Pediatr. 1996;129(1):42-49. 82. Wheeler KA, West RJ, Lloyd JK, Barley J. Double blind trial of bezafibrate in familial hypercholesterolaemia. Arch Dis Child. 1985;60(1):34-37. 83. Wiegman A, de Groot E, Hutten BA. Arterial intima-media thickness in children heterozygous for familial hypercholesterolaemia. Lancet. 2004;363(9406):369-370. 84. Lombardi MP, Redeker EJ, Defesche JC, et al. Molecular genetic testing for familial hypercholesterolemia: spectrum of LDL receptor gene mutations in The Netherlands. Clin Genet. 2000;57(2):116-124. 85. Civeira F, International Panel on Management of Familial Hypercholesterolemia. Guidelines for the diagnosis and management of heterozygous familial hypercholesterolemia. Atherosclerosis. 2004;173(1):55-68. 86. Thorsson B, Sigurdsson G, Gudnason V. Systematic family screening for familial hypercholesterolemia in Iceland. Arterioscler Thromb Vasc Biol. 2003;23(2):335-338.
38 87. Kotze MJ, Langenhoven E, Theart L, Marx MP, Oosthuizen CJ. Report on a molecular diagnostic service for familial hypercholesterolemia in Afrikaners. Genet Couns. 1994;5(1):15-21. 88. Wiegman A, Rodenburg J, de Jongh S, et al. Family history and cardiovascular risk in familial hypercholesterolemia: data in more than 1000 children. Circulation. 2003;107(11):1473-1478. 89. Koivisto PV, Koivisto UM, Miettinen TA, Kontula K. Diagnosis of heterozygous familial hypercholesterolemia. DNA analysis complements clinical examination and analysis of serum lipid levels. Arterioscler & Thromb. 1992;12(5):584-592. 90. Laskarzewski PM, Khoury P, Kelly K, Mellies MJ, Morrison JA, Glueck CJ. Prevalence of familial hypertriglyceridemia: the Princeton School District Family Study. Prev Med. 1982;11(3):317-345. 91. Williams RR, Hunt SC, Schumacher MC, et al. Diagnosing heterozygous familial hypercholesterolemia using new practical criteria validated by molecular genetics. Am J Cardiol. 1993;72(2):171-176. 92. Bao W, Srinivasan SR, Wattigney WA, Berenson GS. Usefulness of childhood low-density lipoprotein cholesterol level in predicting adult dyslipidemia and other cardiovascular risks. The Bogalusa Heart Study. Arch Intern Med. 1996;156(12):1315-1320. 93. Berenson GS, Srinivasan SR, Frerichs RR, Webber LS. Serum high density lipoprotein and its relationship to cardiovascular disease risk factor variables in children--the Bogalusa heart study. Lipids. 1979;14(1):91-98. 94. Clarke WR, H.G. S, Leaverton PE, Conner WE, Lauer RM. Tracking of blood lipids and blood pressures in school age children: the Muscatine study. Circulation. 1978;58(4):626-634. 95. Lauer RM, Lee J, Clarke WR. Predicting adult cholesterol levels from measurements in childhood and adolescence: the Muscatine Study. Bull N Y Acad Med. 1989;65(10):1127-1142; discussion 1154-1160. 96. Stuhldreher WL, Orchard TJ, Donahue RP, Kuller LH, Gloninger MF, Drash AL. Cholesterol screening in childhood: sixteen-year Beaver County Lipid Study experience. J Pediatr. 1991;119(4):551-556. 97. Namboodiri KK, Green PP, Walden C, et al. The Collaborative Lipid Research Clinics Program Family Study. II. Response rates, representativeness of the sample, and stability of lipid and lipoprotein levels. Am J Epidemiol. 1984;119(6):944-958. 98. Laskarzewski P, Morrison JA, de Groot I, et al. Lipid and lipoprotein tracking in 108 children over a four year period. Pediatrics. 1979;64(5):584-591. 99. Guo S, Beckett L, Chumlea WC, Roche AF, Siervogel RM. Serial analysis of plasma lipids and lipoproteins from individuals 9-21 y of age. Am J Clin Nutr. 1993;58(1):61-67. 100. Baumgartner RN, Guo S, Roche AF. Tracking of lipids and lipoproteins in adolescents from 12-22 years of age. The Fels longitudinal study. Ann N Y Acad Sci. 1991;623:406-409. 101. Twisk JW, Kemper HC, van Mechelen W, Post GB. Tracking of risk factors for coronary heart disease over a 14-year period: a comparison between lifestyle and biologic risk factors with data from the Amsterdam Growth and Health Study. Am J Epidemiol. 1997;145(10):888-898. 102. Mohler B, Ackermann-Liebrich U, Steffen T, Staehelin HB. Cholesterol screening in childhood: results of a 9-year follow-up study in Swiss and Italian children in Switzerland. Soz Praventivmed. 1996;41(6):333-340. 103. Vobecky JS, David P, Vobecky J. Dietary habits in relation to tracking of cholesterol level in young adolescents: a nine-year follow-up. Ann Nutr Metab. 1988;32(5-6):312-323. 104. Katzmarzyk PT, Perusse L, Malina RM, Bergeron J, Despres JP, Bouchard C. Stability of indicators of the metabolic syndrome from childhood and adolescence to young adulthood: the Quebec Family Study. J Clin Epidemiol. 2001;54(2):190-195. 105. Kallio MJ, Salmenpera L, Siimes MA, Perheentupa J, Miettinen TA. Tracking of serum cholesterol and lipoprotein levels from the first year of life. Pediatrics. 1993;91(5):949-954. 106. Bastida S, Sanchez-Muniz FJ, Cuena R, Perea S, Aragones A. High density lipoprotein-cholesterol changes in children with high cholesterol levels at birth. Eur J Pediatr. 2002;161(2):94-98. 107. Porkka KV, Viikari JS, Akerblom HK. Tracking of serum HDL-cholesterol and other lipids in children and adolescents: the Cardiovascular Risk in Young Finns Study. Prev Med. 1991;20(6):713-724.
39 108. Porkka KV, Viikari JS, Taimela S, Dahl M, Akerblom HK. Tracking and predictiveness of serum lipid and lipoprotein measurements in childhood: a 12-year follow-up. The Cardiovascular Risk in Young Finns study. Am J Epidemiol. 1994;140(12):1096-1110. 109. Fuentes RM, Notkola IL, Shemeikka S, Tuomilehto J, Nissinen A. Tracking of serum total cholesterol during childhood: an 8-year follow-up population-based family study in eastern Finland. Acta Paediatr. 2003;92(4):420-424. 110. Eisenmann JC, Welk GJ, Wickel EE, Blair SN, Aerobics Center Longitudinal S. Stability of variables associated with the metabolic syndrome from adolescence to adulthood: the Aerobics Center Longitudinal Study. Am J Hum Biol. 2004;16(6):690-696. 111. Nicklas TA, von Duvillard SP, Berenson GS. Tracking of serum lipids and lipoproteins from childhood to dyslipidemia in adults: the Bogalusa Heart Study. Int J Sports Med. 2002;23(Suppl 1):S39-43. 112. Orchard TJ, Donahue RP, Kuller LH, Hodge PN, Drash AL. Cholesterol screening in childhood: does it predict adult hypercholesterolemia? The Beaver County experience. J Pediatr. 1983;103(5):687-691. 113. Lauer RM, Clarke WR. Use of cholesterol measurements in childhood for the prediction of adult hypercholesterolemia. The Muscatine Study. JAMA. 1990;264(23):3034-3038. 114. Barnett AG, van der Pols JC, Dobson AJ. Regression to the mean: what it is and how to deal with it. Int J Epidemiol. 2004;34:215-220. 115. Bell MM, Joseph S. Screening 1140 fifth graders for hypercholesterolemia: family history inadequate to predict results. J Am Board Fam Pract. 1990;3(4):259-263. 116. Davidson DM, Van Camp J, Iftner CA, Landry SM, Bradley BJ, Wong ND. Family history fails to detect the majority of children with high capillary blood total cholesterol. J Sch Health. 1991;61(2):75-80. 117. Diller PM, Huster GA, Leach AD, Laskarzewski PM, Sprecher DL. Definition and application of the discretionary screening indicators according to the National Cholesterol Education Program for Children and Adolescents. J Pediatr. 1995;126(3):345-352. 118. Gagliano NJ, Emans SJ, Woods ER. Cholesterol screening in the adolescent. J Adolesc Health. 1993;14(2):104-108. 119. Griffin TC, Christoffel KK, Binns HJ, McGuire PA. Family history evaluation as a predictive screen for childhood hypercholesterolemia. Pediatric Practice Research Group. Pediatrics. 1989;84(2):365-373. 120. Muhonen LE, Burns TL, Nelson RP, Lauer RM. Coronary risk factors in adolescents related to their knowledge of familial coronary heart disease and hypercholesterolemia: the Muscatine Study. Pediatrics. 1994;93(3):444-451. 121. Primrose ED, Savage JM, Boreham CA, Cran GW, Strain JJ. Cholesterol screening and family history of vascular disease. Arch Dis Child. 1994;71(3):239-242. 122. Resnicow K, Cross D, Lacosse J, Nichols P. Evaluation of a school-site cardiovascular risk factor screening intervention. Prev Med. 1993;22(6):838-856. 123. Sanchez Bayle M, Gonzalez Vergaz A, Garcia Cuartero B, Santos Tapia M, Gonzalez Requejo A. Is a parental history of coronary arterial disease in children as discriminating as their lipoprotein profile? Nino Jesus Group. Int J Cardiol. 1992;36(3):267-271. 124. Steiner NJ, Neinstein LS, Pennbridge J. Hypercholesterolemia in adolescents: effectiveness of screening strategies based on selected risk factors. Pediatrics. 1991;88(2):269-275. 125. Troxler RG, Park MK, Miller MA, Karnavas BA, Lee DH. Predictive value of family history in detecting hypercholesterolemia in predominantly Hispanic adolescents. Tex Med. 1991;87(11):75- 79. 126. Dennison BA, Kikuchi DA, Srinivasan SR, Webber LS, Berenson GS. Parental history of cardiovascular disease as an indication for screening for lipoprotein abnormalities in children. J Pediatr. 1989;115(2):186-194. 127. O'Loughlin J, Lauzon B, Paradis G, et al. Usefulness of the American Academy of Pediatrics recommendations for identifying youths with hypercholesterolemia. Pediatrics. 2004;113(6):1723-1727. 128. Wadowski SJ, Karp RJ, Murray-Bachmann R, Senft C. Family history of coronary artery disease and cholesterol: screening children in a disadvantaged inner-city population. Pediatrics. 1994;93(1):109-113.
40 129. Washington RL, Harrigan L, Day M. Early experience in treating hyperlipidemia in children. Ann N Y Acad Sci. 1991;623:434. 130. Lannon CM, Earp J. Parents' behavior and attitudes toward screening children for high serum cholesterol levels. Pediatrics. 7/7/04 1992;89(6 Pt 2):1159-1163. 131. Boulton TJ. The validity of screening for hypercholesterolaemia at different ages from 2 to 17 years. Aust N Z J Med. 1979;9(5):542-546. 132. Williams RR, Hunt SC, Barlow GK, et al. Prevention of familial cardiovascular disease by screening for family history and lipids in youths. Clin Chem. 1992;38(8B Pt 2):1555-1560. 133. Dennison BA, Jenkins PL, Pearson TA. Challenges to implementing the current pediatric cholesterol screening guidelines into practice. Pediatrics. 1994;94(3):296-302. 134. Berenson GS, Srinivasan S. Cholesterol as a risk factor for early atherosclerosis: the Bogalusa Heart Study. Prog Pediatr Cardiol. 2003;17:113-122. 135. Andersen LB, Wedderkopp N, Hansen HS, Cooper AR, Froberg K. Biological cardiovascular risk factors cluster in Danish children and adolescents: the European Youth Heart Study. Prev Med. 2003;37(4):363-367. 136. Bergstrom E, Hernell O, Persson LA. Endurance running performance in relation to cardiovascular risk indicators in adolescents. Int J Sports Med. 1997;18(4):300-307. 137. Bonora E, Zenere M, Branzi P, et al. Influence of body fat and its regional localization on risk factors for atherosclerosis in young men. Am J Epidemiol. 1992;135(11):1271-1278. 138. Bonora E, Targher G, Branzi P, et al. Cardiovascular risk profile in 38-year and 18-year-old men. Contribution of body fat content and regional fat distribution. Int J Obes Relat Metab Disord. 1996;20(1):28-36. 139. Demerath E, Muratova V, Spangler E, Li J, Minor VE, Neal WA. School-based obesity screening in rural Appalachia. Prev Med. 2003;37(6 Pt 1):553-560. 140. DeStefano F, Berg RL, Griese GG, Jr. Determinants of serum lipid and lipoprotein concentrations in children. Epidemiology. 1995;6(4):446-449. 141. Douglas MB, Birrer RB, Medidi S, Schlussel YR. Obese children should be screened for hypercholesterolemia. J Health Care Poor Underserved. 1996;7(1):24-35. 142. Durant RH, Linder CW, Jay S, Harkness JW, Gray RG. The influence of a family history of CHD risk factors on serum lipoprotein levels in black children and adolescents. J Adolesc Health Care. 1982;3(2):75-81. 143. Durant RH, Linder CW, Mahoney OM. Relationship between habitual physical activity and serum lipoprotein levels in white male adolescents. J Adolesc Health Care. 1983;4(4):235-240. 144. Durant RH, Baranowski T, Rhodes T, et al. Association among serum lipid and lipoprotein concentrations and physical activity, physical fitness, and body composition in young children. J Pediatr. 1993;123(2):185-192. 145. Dwyer T, Gibbons LE. The Australian Schools Health and Fitness Survey. Physical fitness related to blood pressure but not lipoproteins. Circulation. 1994;89(4):1539-1544. 146. Eisenmann JC, Malina RM. Age-related changes in subcutaneous adipose tissue of adolescent distance runners and association with blood lipoproteins. Ann Hum Biol. 2002;29(4):389-397. 147. Fripp RR, Hodgson JL, Kwiterovich PO, Werner JC, Schuler HG, Whitman V. Aerobic capacity, obesity, and atherosclerotic risk factors in male adolescents. Pediatrics. 1985;75(5):813-818. 148. Giovannini M, Bellu R, Ortisi MT, Incerti P, Riva E. Cholesterol and lipoprotein levels in Milanese children: relation to nutritional and familial factors. J Am Coll Nutr. 1992;11(Suppl):28S-31S. 149. Glassman MS, Schwarz SM. Cholesterol screening in children: should obesity be a risk factor? J Am Coll Nutr. 1993;12(3):270-273. 150. Howard JK, Bindler RM, Dimico GS, et al. Cardiovascular risk factors in children: a Bloomsday research report. J of Pediatr Nurs. 1991;6(4):222-229. 151. Jarvisalo MJ, Jartti L, Nanto-Salonen K, et al. Increased Aortic Intima-Media Thickness: A Marker of Preclinical Atherosclerosis in High-Risk Children. Circulation. 2001;104(24):2943- 2947. 152. Larsson B, Vaara I. Cholesterol screening of seven-year-old children. How to identify children at risk. Acta Paediatr. 1992;81(4):315-318. 153. Kunz F, Pechlaner C, Hortnagl H, Pfister R. The smoker's paradox and the real risk of smoking. Eur J Epidemiol 2005;20(2):161-167.
41 154. Kwiterovich PO, Jr., Barton BA, McMahon RP, et al. Effects of diet and sexual maturation on low-density lipoprotein cholesterol during puberty: the Dietary Intervention Study in Children (DISC). Circulation. 1997;96(8):2526-2533. 155. Macek M, Bell D, Rutenfranz J, et al. A comparison of coronary risk factors in groups of trained and untrained adolescents. Eur J Appl Physiol Occup Physiol. 1989;58(6):577-582. 156. Marti B, Vartiainen E. Relation between leisure time exercise and cardiovascular risk factors among 15-year-olds in eastern Finland. J Epidemiol Community Health. 1989;43(3):228-233. 157. Raitakari OT, Taimela S, Porkka KV, Viikari JS. Effect of leisure-time physical activity change on high-density lipoprotein cholesterol in adolescents and young adults. Ann Med. 1996;28(3):259- 263. 158. Resnicow K, Futterman R, Vaughan RD. Body mass index as a predictor of systolic blood pressure in a multiracial sample of US schoolchildren. Ethn Dis. 1993;3(4):351-361. 159. Ribeiro JC, Guerra S, Oliveira J, et al. Physical activity and biological risk factors clustering in pediatric population. Prev Med. 2004;39(3):596-601. 160. Shear CL. The relationship between parental history of vascular disease and cardiovascular disease risk factors in children: the Bogalusa Heart Study. Am J Epidemiol. 1985;122(5):762-771. 161. Simon JA, Morrison JA, Similo SL, McMahon RP, Schreiber GB. Correlates of high-density lipoprotein cholesterol in Black girls and White girls: the NHLBI Growth and Health Study. Am J Pub Health. 1995;85(12):1698-1702. 162. Suter E, Hawes MR. Relationship of physical activity, body fat, diet, and blood lipid profile in youths 10-15 yr. Med Sci Sports Exerc. 1993;25(6):748-754. 163. Thorland WG, Gilliam TB. Comparison of serum lipids between habitually high and low active pre-adolescent males. Med Sci Sports Exerc. 1981;13(5):316-321. 164. Tolfrey K, Campbell IG, Jones AM. Selected predictor variables and the lipid-lipoprotein profile of prepubertal girls and boys. Med Sci Sports Exerc. 1999;31(11):1550-1557. 165. Tonstad S, Leren TP, Sivertsen M, Ose L. Determinants of lipid levels among children with heterozygous familial hypercholesterolemia in Norway. Arterioscler Thromb Vasc Biol. 1995;15(8):1009-1014. 166. Twisk JW, Kemper HC, Van Mechelen W, Post GB. Clustering of risk factors for coronary heart disease. the longitudinal relationship with lifestyle. Ann Epidemiol. 2001;11(3):157-165. 167. Twisk JW, Kemper HC, van Mechelen W, Post GB, van Lenthe FJ. Body fatness: longitudinal relationship of body mass index and the sum of skinfolds with other risk factors for coronary heart disease. Int J Obes Relat Metab Disord. 1998;22(9):915-922. 168. Twisk JW, Kemper HC, Mellenbergh GJ, van Mechelen W, Post GB. Relation between the longitudinal development of lipoprotein levels and lifestyle parameters during adolescence and young adulthood. Ann Epidemiol. 1996;6(3):246-256. 169. van Lenthe FJ, van Mechelen W, Kemper HC, Twisk JW. Association of a central pattern of body fat with blood pressure and lipoproteins from adolescence into adulthood. The Amsterdam Growth and Health Study. Am J Epidemiol. 1998;147(7):686-693. 170. van Stiphout WA, Hofman A, de Bruijn AM, Valkenburg HA. Distributions and determinants of total and high-density lipoprotein cholesterol in Dutch children and young adults. Prev Med. 1985;14(2):169-180. 171. Ward SD, Melin JR, Lloyd FP, Norton JA, Jr., Christian JC. Determinants of plasma cholesterol in children--a family study. Am J Clin Nutr. 1980;33(1):63-70. 172. Wong ND, Hei TK, Qaqundah PY, Davidson DM, Bassin SL, Gold KV. Television viewing and pediatric hypercholesterolemia. Pediatrics. 1992;90(1 Pt 1):75-79. 173. Jiang X, Srinivasan SR, Webber LS, Wattigney WA, Berenson GS. Association of fasting insulin level with serum lipid and lipoprotein levels in children, adolescents, and young adults: the Bogalusa Heart Study. Arch Intern Med. 1995;155(2):190-196. 174. Gliksman MD, Lazarus R, Wilson A. Differences in serum lipids in Australian children: is diet responsible? Int J Epidemiol. 1993;22(2):247-254. 175. Bachman RP, Schoen EJ, Stembridge A, Jurecki ER, Imagire RS. Compliance with childhood cholesterol screening among members of a prepaid health plan. Am J Dis Child. 1993;147(4):382- 385. 176. Bistritzer T, Rosenzweig L, Barr J, et al. Lipid profile with paternal history of coronary heart disease before age 40. Arch Dis Child. 1995;73(1):62-65.
42 177. Davidson DM, Bradley BJ, Landry SM, Iftner CA, Bramblett SN. School-based blood cholesterol screening. J Pediatr Health Care. 1989;3(1):3-8. 178. Faigel HC. Screening college students for hypercholesterolemia. J Am Coll Health. 1992;40(6):272-275. 179. Garcia RE, Moodie DS. Routine cholesterol surveillance in childhood. Pediatrics. 1989;84(5):751-755. 180. Hammond J, Chinn S, Richardson H, Rona R. Response to venepuncture for monitoring in primary schools. Arch Dis Child. 1994;70(5):367-369; discussion 369-372. 181. Heyden S, Schneider KA, Roberts KF. Effectiveness of education-screening on cholesterol levels of students. Ann Nutr Metab. 1991;35(2):71-76. 182. Lansing AM, Barbie RN, Shaheen KA. Regional cholesterol screening: problems and potentials. J Ky Med Assoc. 1990;88(4):170-174. 183. Manchester RA, McDuffie C, Diamond E. Screening for hypercholesterolemia in college students. J Am Coll Health. 1989;37(4):149-153. 184. Polonsky SM, Bellet PS, Sprecher DL. Primary hyperlipidemia in a pediatric population: classification and effect of dietary treatment. Pediatrics. 1993;91(1):92-96. 185. Skovby F, Micic S, Jepsen B, et al. Screening for familial hypercholesterolaemia by measurement of apolipoproteins in capillary blood. Arch Dis Child. 1991;66(7):844-847. 186. Goff DC, Jr., Donker GA, Ragan JD, Jr., et al. Cholesterol screening in pediatric practice. Pediatrics. 1991;88(2):250-258. 187. Nader PR, Yang M, Luepker RV, et al. Parent and physician response to children's cholesterol values of 200 mg/dL or greater: the Child and Adolescent Trial for Cardiovascular Health Experiment. Pediatrics. 1997;99(5):E5. 188. Sinaiko AR, Prineas RJ. Reduction of cardiovascular disease: what is the role of the pediatrician? Pediatrics. 1998;102(5):E61. 189. Bastida S, Perea S, Sanchez-Muniz FJ. Do neonates with high serum cholesterol levels have a different high density lipoprotein composition? Eur J Pediatr. 1998;157(1):66-70. 190. Andersen GE, Lous P, Friis-Hansen B. Screening for hyperlipoproteinemia in 10,000 Danish newborns. Follow-up studies in 522 children with elevated cord serum VLDL-LDL-cholesterol. Acta Paediatr Scand. 1979;68(4):541-545. 191. Kenny D, Ward OC, Mulhern B, Rashied AA. Failure of cord serum cholesterol and betalipoprotein as screening tests for familial hyperlipoproteinaemia. Ir J Med Sci. 1984;153(3):100-105. 192. Cress HR, Shaher RM, Greenberg MH, Laffin R. Lipoproteins in neonates. J Pediatr. 1974;84(4):585-587. 193. Barnes K, Nestel PJ, Pryke ES, Whyte HM. Neonatal plasma lipids. Med J Austr. 1972;2(18):1002-1005. 194. Andersen GE, Brokhattingen K, Lous P. Familial hypobetalipoproteinaemia in 9 children diagnosed as the result of cord blood screening for hypolipoproteinaemia in 10 000 Danish newborns. Arch Dis Child. 1979;54(9):691-694. 195. Goldstein JL, Albers JJ, Schrott HG, Hazzard WR, Bierman EL, Motulsky AG. Plasma lipid levels and coronary heart disease in adult relatives of newborns with normal and elevated cord blood lipids. Am J Hum Genet. 1974;26(6):727-735. 196. Darmady JM, Fosbrooke AS, Lloyd JK. Prospective study of serum cholesterol levels during first year of life. Br Med J. 1972;2(815):685-688. 197. Tsang RC, Fallat RW, Glueck CJ. Cholesterol at birth and age 1: comparison of normal and hypercholesterolemic neonates. Pediatrics. 1974;53(4):458-470. 198. Kwiterovich PO, Jr., Levy RI, Fredrickson DS. Neonatal diagnosis of familial type-II hyperlipoproteinaemia. Lancet. 1973;1(7795):118-121. 199. Andersen GE, Nielsen HG. Neonatal screening for hyperlipoproteinemia. Methods for direct estimation of cord serum VLDL + LDL. Clin Chim Acta. 1976;66(1):29-41. 200. Van Biervliet JP, Vercaemst R, De Keersgieter W, Vinaimont N, Caster H, Rosseneu M. Evolution of lipoprotein patterns in newborns. Acta Paediatr Scand. 1980;69(5):593-596. 201. Brewster TG, Waite DJ, Hudson GA. Quantitation of beta-lipoprotein in cord serum by rate nephelometric immunoassay: a potential screening test for familial hypercholesterolemia. Clin Chem. 1982;28(5):1192-1196.
43 202. Wood WG, Schumacher M, Weigert S. (Apo)lipoprotein(a) concentrations at birth and in the first days and months of life--studies on the distribution of serum levels and the predictive value of measurements made at this time. Eur J Clin Chem Clin Biochem. 1995;33(3):139-145. 203. Vladutiu GD, Glueck CJ, Schultz MT, McNeely S, Guthrie R. beta-Lipoprotein quantitation in cord blood spotted on filter paper: a screening test. Clin Chem. 1980;26(9):1285-1290. 204. Wilcken DE, Blades BL, Dudman NP. A neonatal screening approach to the detection of familial hypercholesterolaemia and family-based coronary prevention. J Inherit Metab Dis. 1988;11(Suppl 1):87-90. 205. Blades BL, Dudman NP, Wilcken DE. Screening for familial hypercholesterolemia in 5000 neonates: a recall study. Pediatr Res. 1988;23(5):500-504. 206. Bangert SK, Eldridge PH, Peters TJ. Neonatal screening for familial hypercholesterolaemia by immunoturbidimetric assay of apolipoprotein B in dried blood spots. Clin Chim Acta. 1992;213(1- 3):95-101. 207. Van Biervliet JP, Vinaimont N, Caster H, Rosseneu M, Belpaire F. A screening procedure for dyslipoproteinemia in the newborn. Apoprotein quantitation on dried blood spots. Clin Chim Acta. 1982;120(2):191-200. 208. Roseneu M, Van Biervliet JP. Screening and follow-up of infants with dyslipoproteinemia. Prog Clin Biol Res. 1985;188:79-86. 209. Beeso J, Wong N, Ayling R, et al. Screening for hypercholesterolaemia in 10,000 neonates in a multi-ethnic population. Eur J Pediatr. 1999;158(10):833-837. 210. Vuorio AF, Turtola H, Kontula K. Neonatal diagnosis of familial hypercholesterolemia in newborns born to a parent with a molecularly defined heterozygous familial hypercholesterolemia. Arterioscler Thromb Vasc Biol. 1997;17(11):3332-3337. 211. Croyle RT, Sun YC, Louie DH. Psychological minimization of cholesterol test results: moderators of appraisal in college students and community residents. Health Psychol. 1993;12(6):503-507. 212. McHale SM, Tershakovec AM, Corneal DA, Tournier BA, Shannon BM. Psychosocial factors in nutrition education for hypercholesterolemic children. Ann Behav Med. 1998;20(3):233-240. 213. Rosenberg E, Lamping DL, Joseph L, Pless IB, Franco ED. Cholesterol screening of children at high risk: behavioural and psychological effects. Can Med Assoc J. 1997;156(4):489-496. 214. Tonstad S. Familial hypercholesterolaemia: a pilot study of parents' and children's concerns. Acta Paediatr. 1996;85(11):1307-1313. 215. Rosenthal SL, Knauer-Black S, Stahl MP, Catalanotto TJ, Sprecher DL. The psychological functioning of children with hypercholesterolemia and their families. A preliminary investigation. Clinical Pediatrics. 1993;32(3):135-141. 216. Hanna KJ, Ewart CK, Kwiterovich PO, Jr. Child problem solving competence, behavioral adjustment and adherence to lipid-lowering diet. Patient Educ Couns. 1990;16(2):119-131. 217. Epstein LH, Kuller LH, Wing RR, Valoski A, McCurley J. The effect of weight control on lipid changes in obese children. Am J Dis Child. 1989;143(4):454-457. 218. Obarzanek E, Kimm SY, Barton BA, et al. Long-term safety and efficacy of a cholesterol- lowering diet in children with elevated low-density lipoprotein cholesterol: seven-year results of the Dietary Intervention Study in Children (DISC). Pediatrics. 2001;107(2):256-264. 219. DISC Collaborative Research Group. Efficacy and safety of lowering dietary intake of fat and cholesterol in children with elevated low-density lipoprotein cholesterol. The Dietary Intervention Study in Children (DISC). The Writing Group for the DISC Collaborative Research Group. JAMA. 1995;273(18):1429-1435. 220. Shannon BM, Tershakovec AM, Martel JK, Achterberg CL, Cortner JA. Reduction of elevated LDL-cholesterol levels of 4- to 10-year-old children through home-based dietary education. Pediatrics. 1994;94(6):923-927. 221. Becque MD, Katch VL, Rocchini AP, Marks CR, Moorehead C. Coronary risk incidence of obese adolescents: reduction by exercise plus diet intervention. Pediatrics. 1988;81(5):605-612. 222. Walter HJ, Hofman A, Connelly PA, Barrett LT, Kost KL. Primary prevention of chronic disease in childhood: changes in risk factors after one year of intervention. Am J Epidemiol. 1985;122(5):772-781. 223. Dennison BA, Levine DM. Randomized, double-blind, placebo-controlled, two-period crossover clinical trial of psyllium fiber in children with hypercholesterolemia. J Pediatr. 1993;123(1):24- 29.
44 224. Engler MM, Engler MB, Malloy MJ, et al. Antioxidant vitamins C and E improve endothelial function in children with hyperlipidemia: Endothelial Assessment of Risk from Lipids in Youth (EARLY) Trial. Circulation. 2003;108(9):1059-1063. 225. Ferguson MA, Gutin B, Le NA, et al. Effects of exercise training and its cessation on components of the insulin resistance syndrome in obese children. Int J Obes Relat Metab Disord. 1999;23(8):889-895. 226. Kang HS, Gutin B, Barbeau P, et al. Physical training improves insulin resistance syndrome markers in obese adolescents. Med Sci Sports Exerc. 2002;34(12):1920-1927. 227. Kuehl KS, Cockerham JT, Hitchings M, Slater D, Nixon G, Rifai N. Effective control of hypercholesterolemia in children with dietary interventions based in pediatric practice. Prev Med. 1993;22(2):154-166. 228. Stergioulas A, Tripolitsioti A, Messinis D, Bouloukos A, Nounopoulos C. The effects of endurance training on selected coronary risk factors in children. Acta Paediatr. 1998;87(4):401- 404. 229. Williams CL, Bollella M, Spark A, Puder D. Soluble fiber enhances the hypocholesterolemic effect of the step I diet in childhood. J Am Coll Nutr. 1995;14(3):251-257. 230. Boreham CA, Wallace WF, Nevill A. Training effects of accumulated daily stair-climbing exercise in previously sedentary young women. Prev Med. 2000;30(4):277-281. 231. Gold K, Wong N, Tong A, et al. Serum apolipoprotein and lipid profile effects of an oat-bran- supplemented, low-fat diet in children with elevated serum cholesterol. Ann N Y Acad Sci. 1991;623:429-431. 232. Linder CW, DuRant RH, Mahoney OM. The effect of physical conditioning on serum lipids and lipoproteins in white male adolescents. Med Sci Sports Exerc. 1983;15(3):232-236. 233. Savage MP, Petratis MM, Thomson WH, Berg K, Smith JL, Sady SP. Exercise training effects on serum lipids of prepubescent boys and adult men. Med Sci Sports Exerc. 1986;18(2):197-204. 234. Engler MM, Engler MB, Malloy M, et al. Docosahexaenoic acid restores endothelial function in children with hyperlipidemia: results from the EARLY study. Int J Clin Pharmacol Ther. 2004;42(12):672-679. 235. McCrindle BW, Helden E, Conner WT. Garlic extract therapy in children with hypercholesterolemia. Arch Pediatr Adolesc Med. 1998;152(11):1089-1094. 236. Engler MM, Engler MB, Malloy MJ, Paul SM, Kulkarni KR, Mietus-Snyder ML. Effect of docosahexaenoic acid on lipoprotein subclasses in hyperlipidemic children (the EARLY study). Am J Cardiol. 2005;95(7):869-871. 237. Fernandes J, Dijkhuis-Stoffelsma R, Groot PH, Grose WF, Ambagtsheer JJ. The effect of a virtually cholesterol-free, high-linoleic-acid vegetarian diet on serum lipoproteins of children with familial hypercholesterolemia (type II-A). Acta Paediatr Scand. 1981;70(5):677-682. 238. Tonstad S, Refsum H, Sivertsen M, Christophersen B, Ose L, Ueland PM. Relation of total homocysteine and lipid levels in children to premature cardiovascular death in male relatives. Pediatr Res. 1996;40(1):47-52. 239. American Heart Association Committee Report, Glueck CJ, McGill HC, Jr., Shank RE, Lauer RM. Value and safety of diet modification to control hyperlipidemia in childhood and adolescence. Circulation. 1978;58:381A-390. 240. Hedman M, Neuvonen PJ, Neuvonen M, Antikainen M. Pharmacokinetics and pharmacodynamics of pravastatin in children with familial hypercholesterolemia. Clin Pharmacol Ther. 2003;74(2):178-185. 241. Dirisamer A, Hachemian N, Bucek RA, Wolf F, Reiter M, Widhalm K. The effect of low-dose simvastatin in children with familial hypercholesterolaemia: a 1-year observation. Eur J Pediatr. 2003;162(6):421-425. 242. Sinzinger H, O'Grady J. Professional athletes suffering from familial hypercholesterolaemia rarely tolerate statin treatment because of muscular problems. Br J Clin Pharmacol. 2004;57(4):525-528. 243. Ducobu J, Brasseur D, Chaudron JM, et al. Simvastatin use in children. Lancet. 1992;339(8807):1488. 244. Stefanutti C, Lucani G, Vivenzio A, Di Giacomo S. Diet only and diet plus simvastatin in the treatment of heterozygous familial hypercholesterolemia in childhood. Drugs Exp Clin Res. 1999;25(1):23-28.
45 245. Curtis DM, Driscoll DJ, Goldman DH, Weidman WH. Loss of dental enamel in a patient taking cholestyramine. Mayo Clin Proc. 1991;66(11):1131. 246. Farah JR, Kwiterovich PO, Neill CA. A study of the dose-effect relationship of cholestyramine in children with familial hypercholesterolemia. Adv Exp Med Biol. 1977;82:212-215. 247. Farah JR, Kwiterovich PO, Jr., Neill CA. Dose-effect relation of cholestryamine in children and young adults with familial hypercholesterolaemia. Lancet. 1977;1(8002):59-63. 248. Groot PH, Dijkhuis-Stoffelsma R, Grose WF, Ambagtsheer JJ, Fernandes J. The effects of colestipol hydrochloride on serum lipoprotein lipid and apolipoprotein B and A-I concentrations in children heterozygous for familial hypercholesterolemia. Acta Paediatr Scand. 1983;72(1):81-85. 249. Hansen D, Michaelsen KF, Skovby F. Growth during treatment of familial hypercholesterolemia. Acta Paediatr. 1992;81(12):1023-1025. 250. Harvengt C, Desager JP. Colestipol in familial type II hyperlipoproteinemia: a three-year trial. Clin Pharmacol Ther. 1976;20(3):310-314. 251. Koletzko B, Kupke I, Wendel U. Treatment of hypercholesterolemia in children and adolescents. Acta Paediatr. 1992;81(9):682-685. 252. Liacouras CA, Coates PM, Gallagher PR, Cortner JA. Use of cholestyramine in the treatment of children with familial combined hyperlipidemia. J Pediatr. 1993;122(3):477-482. 253. Schwarz KB, Goldstein PD, Witztum JL, Schonfeld G. Fat-soluble vitamin concentrations in hypercholesterolemic children treated with colestipol. Pediatrics. 1980;65(2):243-250. 254. Tonstad S, Ose L. Colestipol tablets in adolescents with familial hypercholesterolaemia. Acta Paediatr. 1996;85(9):1080-1082. 255. Tonstad S, Refsum H, Ose L, Ueland PM. The C677T mutation in the methylenetetrahydrofolate reductase gene predisposes to hyperhomocysteinemia in children with familial hypercholesterolemia treated with cholestyramine. J Pediatr. 1998;132(2):365-368. 256. West RJ, Lloyd JK. Use of cholestyramine in treatment of children with familial hypercholesterolaemia. Arch Dis Child. 1973;48(5):370-374. 257. West RJ, Lloyd JK. The effect of cholestyramine on intestinal absorption. Gut. 1975;16(2):93-98. 258. West RJ, Fosbrooke AS, Lloyd JK. Treatment of children with familial hypercholesterolaemia. Postgrad Med J. 1975;51(8):suppl 82-87. 259. West RJ, Lloyd JK, Leonard JV. Long-term follow-up of children with familial hypercholesterolaemia treated with cholestyramine. Lancet. 1980;2(8200):873-875. 260. Glueck CJ, Tsang RC, Fallat RW, Mellies M. Therapy of familial hypercholesterolemia in childhood: diet and cholestyramine resin for 24 to 36 months. Pediatrics. 1977;59(3):433-441. 261. Glueck CJ, Mellies MJ, Dine M, Perry T, Laskarzewski P. Safety and efficacy of long-term diet and diet plus bile acid-binding resin cholesterol-lowering therapy in 73 children heterozygous for familial hypercholesterolemia.[retraction in Glueck CJ, Laskarzewski P, Meelies MJ, Perry T. Pediatrics 1987 Nov;80(5):766]. Pediatrics. 1986;78(2):338-348. 262. Glueck CJ, Tsang RC, Fallat RW, Scheel D. Plasma vitamin A and E levels in children with familial type II hyperlipoproteinemia during therapy with diet and cholestyramine resin. Pediatrics. 1974;54(1):51-55. 263. Glueck CJ, Fallat R, Tsang R. Pediatric familial type II hyperlipoproteinemia: therapy with diet and cholestyramine resin. Pediatrics. 1973;52(5):669-679. 264. Becker M, Staab D, Von Bergman K. Long-term treatment of severe familial hypercholesterolemia in children: effect of sitosterol and bezafibrate. Pediatrics. 1992;89(1):138- 142. 265. Brun LD, Gagne C, Coulombe P, Lupien PJ, Dussault JH, Moorjani S. Effects of dextrothyroxine on the pituitary-thyroid axis in hypercholesterolemic children and goitrous adults. J Clin Endocrinol Metab. 1980;51(6):1306-1310. 266. Colletti RB, Neufeld EJ, Roff NK, McAuliffe TL, Baker AL, Newburger JW. Niacin treatment of hypercholesterolemia in children. Pediatrics. 1993;92(1):78-82. 267. McDuffie JR, Calis KA, Uwaifo GI, et al. Three-month tolerability of orlistat in adolescents with obesity-related comorbid conditions. Obes Res. 2002;10(7):642-650. 268. Stein EA. Treatment of familial hypercholesterolemia with drugs in children. Arteriosclerosis. 1989;9(1 Suppl):I145-151. 269. Steinmetz J, Morin C, Panek E, Siest G, Drouin P. Biological variations in hyperlipidemic children and adolescents treated with fenofibrate. Clin Chim Acta. 1981;112(1):43=53.
46 270. Copperman N, Schebendach J, Arden MR, Jacobson MS. Nutrient quality of fat- and cholesterol- modified diets of children with hyperlipidemia. Arch Pediatr Adol Med. 1995;149(3):333-336. 271. Tonstad S, Novik TS, Vandvik IH. Psychosocial function during treatment for familial hypercholesterolemia. Pediatrics. 1996;98(2 Pt 1):249-255. 272. Segall MM, Tamir I, Fosbrooke AS, Lloyd JK, Wolff OH. Effects of short-term high-carbohydrate feeding on serum triglyceride of children with familial hypercholesterolaemia. Arch Dis Child. 1970;45(241):393-398. 273. Witschi JC, Singer M, Wu-Lee M, Stare FJ. Family cooperation and effectiveness in a cholesterol- lowering diet. J Am Diet Assoc. 1978;72(4):384-389. 274. McKenzie J, Dixon LB, Smiciklas-Wright H, Mitchell D, Shannon B, Tershakovec A. Change in nutrient intakes, number of servings, and contributions of total fat from food groups in 4- to 10- year-old children enrolled in a nutrition education study. J Am Diet Assoc. 1996;96(9):865-873. 275. Moreno LA, Sarria A, Lazaro A, Lasierra MP, Larrad L, Bueno M. Lymphocyte T subset counts in children with hypercholesterolemia receiving dietary therapy. Ann Nutr Metab. 1998;42(5):261- 265. 276. Sanchez-Bayle M, Gonzalez-Requejo A, Baeza J, et al. Diet therapy for hypercholesterolemia in children and adolescents. A follow-up. Arch Pediatr Adol Med. 1994;148(1):28-32. 277. Rose V, Allen DM, Pearse RG, Chapell J. Primary hyperlipoproteinemia in childhood and adolescence: identification and treatment of persons at risk for premature atherosclerosis. Can Med Assoc J. 1976;115(8):753-756, 779. 278. Lifshitz F, Moses N. Growth failure. A complication of dietary treatment of hypercholesterolemia. Am J Dis Child. 1989;143(5):537-542. 279. Cetta F, Driscoll DJ, Lucas AR, et al. Growth patterns of hyperlipidemic children enrolled in a preventive cardiovascular health clinic. Clinical Pediatrics. 1994;33(10):588-592. 280. Jacobson MS, Tomopoulos S, Williams CL, Arden MR, Deckelbaum RJ, Starc TJ. Normal growth in high-risk hyperlipidemic children and adolescents with dietary intervention. Prev Med. 1998;27(6):775-780. 281. Lavigne JV, Brown KM, Gidding S, et al. A cholesterol-lowering diet does not produce adverse psychological effects in children: three-year results from the dietary intervention study in children. Health Psychol. 1999;18(6):604-613. 282. Feoli-Fonseca JC, Levy E, Godard M, Lambert M. Familial lipoprotein lipase deficiency in infancy: clinical, biochemical, and molecular study. J Pediatr. 1998;133(3):417-423. 283. Sanchez-Bayle M, Soriano-Guillen L. Influence of dietary intervention on growth in children with hypercholesterolaemia. Acta Paediatr. 2003;92(9):1043-1046. 284. Tershakovec AM, Jawad AF, Stallings VA, et al. Growth of hypercholesterolemic children completing physician-initiated low-fat dietary intervention. J Pediatr. 1998;133(1):28-34. 285. Kaistha A, Deckelbaum RJ, Starc TJ, Couch SC. Overrestriction of dietary fat intake before formal nutritional counseling in children with hyperlipidemia. Arch Pediatr Adol Med. 2001;155(11):1225-1230. 286. Pugliese MT, et al. Parental health beliefs: a cause of non-organic failure to thrive. Pediatrics. 1987;80(2):175-181. 287. Amundsen AL, Ntanios F, Put N, Ose L. Long-term compliance and changes in plasma lipids, plant sterols and carotenoids in children and parents with FH consuming plant sterol ester- enriched spread. Eur J Clin Nutr. 2004;58(12):1612-1620. 288. Becker M, Staab D, Von Bergmann K. Treatment of severe familial hypercholesterolemia in childhood with sitosterol and sitostanol. J Pediatr. 1993;122(2):292-296. 289. Clarke JT, Cullen-Dean G, Regelink E, Chan L, Rose V. Increased incidence of epistaxis in adolescents with familial hypercholesterolemia treated with fish oil. J Pediatr. 1990;116(1):139- 141. 290. Glassman M, Spark A, Berezin S, Schwartz S, Medow M, Newman LJ. Treatment of type IIa hyperlipidemia in childhood by a simplified American Heart Association diet and fiber supplementation. Am J Dis Child. 1990;144:973-976. 291. Gulesserian T, Widhalm K. Effect of a rapeseed oil substituting diet on serum lipids and lipoproteins in children and adolescents with familial hypercholesterolemia. J Am Coll Nutr. 2002;21(2):103-108.
47 292. Mietus-Snyder M, Malloy MJ. Endothelial dysfunction occurs in children with two genetic hyperlipidemias: improvement with antioxidant vitamin therapy. J Pediatr. 1998;133(1):35-40. 293. Sanchez-Bayle M, Gonzalez-Requejo A, Asensio-Anton J, Ruiz-Jarabo C, Fernandez-Ruiz ML, Baeza J. The effect of fiber supplementation on lipid profile in children with hypercholesterolemia. Clinical Pediatrics. 2001;40(5):291-294. 294. Laurin D, Jacques H, Moorjani S, et al. Effects of a soy-protein beverage on plasma lipoproteins in children with familial hypercholesterolemia. Am J Clin Nutr. 1991;54(1):98-103. 295. Schlierf G, Oster P, Heuck CC, Raetzer H, Schellenberg B. Sitosterol in juvenile type II hyperlipoproteinemia. Atherosclerosis. 1978;30(4):245-248. 296. Zavoral JH, Hannan P, Fields DJ, et al. The hypolipidemic effect of locust bean gum food products in familial hypercholesterolemic adults and children. Am J Clin Nutr. 1983;38(2):285- 294. 297. Tolfrey K, Campbell IG, Batterham AM. Exercise training induced alterations in prepubertal children's lipid-lipoprotein profile. Med Sci Sports Exerc. 1998;30(12):1684-1692. 298. Kavey RE, Kveselis DA, Gaum WE. Valvular and Pediatric Congenital Heart Disease. Am Heart J. 1997;133(2). 299. Tershakovec AM, Brannon SD, Bennett MJ, Shannon BM. The cost of implementation of the Clinical Laboratory Improvement Amendments of 1988--the example of pediatric office-based cholesterol screening. Pediatrics. 1995;96(2 Pt 1):230-234. 300. McKenney JM. An evaluation of cholesterol screening in community pharmacies. Am Pharm. 1993;NS33(7):34-40. 301. Marks D, Wonderling D, Thorogood M, Lambert H, Humphries SE, Neil HA. Cost effectiveness analysis of different approaches of screening for familial hypercholesterolaemia. BMJ. 2002;324(7349):1303. 302. Iannuzzi A, Licenziati MR, Acampora C, al. e. Increased carotid intima-media thickness and stiffness in obese children. Diabetes Care. 2004;27(10):2506-2508. 303. Jarvisalo MJ, Lehtimaki T, Raitakari O. Determinants of arterial nitrate-mediated dilation in children: role of oxidized low-density lipoprotein, endothelial function, and carotid intima-media thickness. Circulation. 2004;109(23):2885-2889. 304. Davis PH, Dawson JD, Riley WA, Lauer RM. Carotid intimal-medial thickness is related to cardiovascular risk factors measured from childhood through middle age: the Muscatine Study. Circulation. 2001;104(23):2815-2819. 305. Rodenburg J, Vissers MN, Wiegman A, Trip MD, Bakker HD, Kastelein JJ. Familial hypercholesterolemia in children. Curr Opin Lipidol. 2004;15(4):405-411. 306. Lavin A, Nauss AH, Newburger JW. A survey of pediatric management of dyslipidemias in New England. Pediatr Cardiol. 1992;13(2):76-79.
48 Figure 1. Defining the Screening Population
Phenotypic Dyslipidemia
Idiopathic Secondary Dyslipidemia Monogenic Dyslipidemia Polygenic, Dyslipidemia Multi-factorial, Secondary to other Undiagnosed Undiagnosed Diagnosed monogenic or diagnosed secondary monogenic syndromes Risk Factor conditions (e.g., dyslipidemia dyslipidemia (e.g., familial diabetes, nephrotic Associated hypercholesterolemia, syndrome) familial defective ApoB)
Children and adolescents identified by screening include those with undiagnosed monogenic dyslipidemia, undiagnosed secondary dyslipidemia, and idiopathic (polygenic, multi-factorial or risk factor associated) dyslipidemia. Children and adolescents with previously known monogenic or secondary dyslipidemia would be specifically evaluated for these indications and are not included in the screening pool for the general population. Figure 2. National Cholesterol Education Program Guidelines for Risk Assessment and Classification of Children
Risk assessment Lipoprotein analysis • 12-hour fast • Measure total cholesterol, HDL-C, triglycerides • Estimate LDL-C‡ Parent total Positive family Do lipoprotein cholesterol >240 history* analysis mg/dL
Acceptable LDL-C Borderline LDL-C High LDL-C Measure total <110 mg/dL 110-129 mg/dL >130 mg/dL cholesterol Repeat cholesterol Repeat lipoprotein analysis and average with previous measurement Acceptable total measurement cholesterol <170 within 5 years mg/dL Provide education on recommended diet and risk factor Acceptable LDL-C Borderline LDL-C High LDL-C reduction† <110 mg/dL 110-129 mg/dL >130 mg/dL
Total cholesterol Repeat total Repeat lipoprotein Risk factor advice Clinical evaluation <170 mg/dL analysis within 5 years (history, physical, lab) Borderline total cholesterol Provide Step I diet and • Secondary causes cholesterol and average Provide education on risk factor • Familial disorders 170-199 mg/dL with previous Total cholesterol recommended diet and interventions† measurement >170 mg/dL risk factor reduction† Intensive clinical Reevaluate status in 1 intervention† year Screen all family High total members Do lipoprotein cholesterol Set target LDL-C analysis >200 mg/dL • Minimal <130 mg/dL • Ideal <110 mg/dL *Positive family history includes parents or grandparents with coronary artery disease, peripheral vascular disease, Step I then Step II diet cerebrovascular disease, or sudden cardiac death at age 55 or younger or parents with total cholesterol ≥240 mg/dL. †Includes smoking cessation, weight loss, exercise, blood pressure control as appropriate. ‡Estimated LDL-C=Total cholesterol – HDL-C – (Triglycerides/5) Figure adapted from Pediatrics, 101(1), American Academy of Pediatrics Committee on Nutrition, Cholesterol in Childhood, 141-147, (1998). Figure 3. Analytic Framework and Key Questions
1
4
Treatment Diet Risk Assessment 5 Reduced incidence and Testing Total cholesterol of adult Exercise dyslipidemia: 6 LDL-C Delayed onset Detection of Total cholesterol and reduced Drug Children and Triglycerides 10 incidence of 2 children and 7 LDL-C adolescents in adolescents Non-HDL-C CHD-related the general with dyslipidemia Combination Triglycerides events population* HDL-C 3 94 8 Non-HDL-C HDL-C Adverse Adverse effects effects
*Includes those without previously known conditions that cause dyslipidemia such as genetic dyslipidemia, diabetes, nephrotic syndrome, organ transplant, and others.
Continued Figure 3. Analytic Framework and Key Questions
Key Questions 1. Is screening for dyslipidemia in children/adolescents effective in delaying the onset and reducing the incidence of CHD- related events? 2. What is the accuracy of screening for dyslipidemia in identifying children/adolescents at increased risk of CHD-related events? 2a. What are abnormal lipid values in children/adolescents? 2b. What are appropriate tests? How well do screening tests (non-fasting total cholesterol, fasting total cholesterol, fasting lipoprotein analysis) identify individuals with dyslipidemia? 2c. How well do lipid levels track from childhood to adulthood? 2d. What is the accuracy of family history in determining risk? 2e. What are other important risk factors? 2f. What are effective screening strategies for children/adolescents (including frequency of testing, optimal age for testing)? 3. What are the adverse effects of screening (including false positives, false negatives, labeling)? 4. In children/adolescents, what is the effectiveness of drug, diet, exercise, and combination therapy in reducing the incidence of adult dyslipidemia, and delaying the onset and reducing the incidence of CHD-related events (including optimal age for initiation of treatment)? 5, 6, 7, 8. What is the effectiveness of drug, diet, exercise or combination therapy for treating dyslipidemia in children/adolescents? 9. What are the adverse effects of drug, diet, exercise, and combination therapy in children/adolescents? 10. Does improving dyslipidemia in childhood reduce the risk of dyslipidemia in adulthood? 11. What are the cost issues involved in screening for dyslipidemia in children/adolescents? Figure 4. Probability Distribution for Heterozygous Familial Hypercholesterolemia, Relative, and the General Population
First-degree relative with FH gene Second-degree relative with FH gene 0.01 50% chance of having FH gene 0.01 25% chance of having FH gene
Non-FH
Non-FH 3/4 Probability 1/2 FH Probability
1/2 FH 1/4
50 100 150 200 250 300 350 400 450 500 550 600 50 100 150 200 250 300 350 400 450 500 550 600 Total Cholesterol Total Cholesterol
Non-FH 0.01 0.01
Third-degree relative with FH gene General Population 12.5% chance of having FH gene Non-FH 0.2% chance of having FH gene 7/8 99.8% Probability Probability
FH 0.2% FH 1/8
50 100 150 200 250 300 350 400 450 500 550 600 50 100 150 200 250 300 350 400 450 500 550 600 Total Cholesterol Total Cholesterol
Continued Figure 4. Probability Distribution for Heterozygous Familial Hypercholesterolemia, Relative, and the General Population
0.01 Probability of FH with a Given Cholesterol Level
For a given cholesterol level of 300 mg/dL, a point probability for FH is indicated by the vertical height Non-FH below that point on the curve for non-FH (N) and FH (F) distributions. The ratio of point probabilities (F/N) estimates the odds that a person with that
Probability cholesterol level comes from the FH distribution or the non-FH distribution. 1/2 FH
1/2 F N
50 100 150 200 250 300 350 400 450 500 550 600
Total Cholesterol
Reprinted from American Journal of Cardiology, 72(2), Williams RR, Hunt SC, Schumacher MC, et al. Diagnosing heterozygous familial hypercholesterolemia using new practical criteria validated by molecular genetics, 171-6, 1993 with permission from Excerpta Medica, Inc. Figure 5a. Studies of Tracking: Total Cholesterol Levels Based on Age at Start of Study
18 21 21 29 30 30 16 29 27 28 36 20 28, 27 14 32 5 4 32 25 25, 26 31 s 12 31 19 19 24 24 3 35 10 7 23 23 18 18 33 34 8 2 22 22
6 17 17 86
Age at Start of Study (year 1
4 16 16
2 14 12 13 9 11 00.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 10 15 Correlation Coefficient
Males & Females Males Females
Continued Figure 5a. Studies of Tracking: Total Cholesterol Levels Based on Age at Start of Study
KEY
Study, year Age at follow-up Study, year Age at follow-up (years) (years) 1. Berenson, 1979 7 22. Lauer, 1989 20-25 2. Berenson, 1979 10 23. Lauer, 1989 20-25 3. Berenson, 1979 13 24. Lauer, 1989 20-25 4. Berenson, 1979 16 25. Lauer, 1989 20-25 5. Twisk, 1997 27-30 26. Lauer, 1989 26-30 6. Mohler, 1996 10 27. Lauer, 1989 20-25 7. Mohler, 1996 14 28. Lauer, 1989 26-30 8. Mohler, 1996 14 29. Lauer, 1989 20-25 9. Kallio, 1993 5 30. Lauer, 1989 26-30 10. Kallio, 1993 5 31. Stuhldreher, 1991 27-30 (mean age 28) 11. Kallio, 1993 5 32. Baumgartner, 1991 < 21 12. Kallio, 1993 5 33. Kelder, 2002 14.1 13. Kallio, 1993 5 34. Kelder, 2002 11.1 14. Kallio, 1993 5 35. Kelder, 2002 14.1 15. Bastida, 2002 4 36. Eisenmann, 2004 15.8 16. Porkka, 1994 15 17. Porkka, 1994 18 18. Porkka, 1994 21 19. Porkka, 1994 24 20. Porkka, 1994 27 21. Porkka, 1994 30 Figure 5b. Studies of Tracking: LDL-C Levels Based on Age at Start of Study
18 18 18 27 26 27 26 16 24 25 25 17 24 14 28 4 22 17 28 23 12 22, 16 16 s 23 3 21 21 10 6 20 20 15 15
of Study (year 8 2 19 19 start 6 14 14
Age at at Age 57 1 4 13 13 2 10 11 9 8 12 00.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Correlation Coeficient
Males & Females Males Females
Continued Figure 5b. Studies of Tracking: LDL-C Levels Based on Age at Start of Study
Key Study, year Age at follow-up Study, year Age at follow-up (years) (years) 1. Berenson, 1979 7 22. Lauer, 1989 20-25 2. Berenson, 1979 10 23. Lauer, 1989 26-30 3. Berenson, 1979 13 24. Lauer, 1989 20-25 4. Berenson, 1979 16 25. Lauer, 1989 26-30 5. Mohler, 1996 10 26. Lauer, 1989 20-25 6. Mohler, 1996 14 27. Lauer, 1989 26-30 7. Mohler, 1996 14 28. Baumgartner, 1991 < 21 8. Kallio, 1993 5 9. Kallio, 1993 5 10. Kallio, 1993 5 11. Kallio, 1993 5 12. Bastida, 2002 4 13. Porkka, 1994 15 14. Porkka, 1994 18 15. Porkka, 1994 21 16. Porkka, 1994 24 17. Porkka, 1994 27 18. Porkka, 1994 30 19. Lauer, 1989 20-25 20. Lauer, 1989 20-25 21. Lauer, 1989 20-25 Figure 5c. Studies of Tracking: HDL-C Levels Based on Age at Start of Study
14 14 22 2323 22
18 29 202121 25 20 13 24 25 16 13 24 19 18 33
s 18 19 14 12 12 17 17 12 32 28 16 2 16 11 11 10 27 26 15 31 15 8 10 1 10 6 3 30 Age at Start of Study (year Study of Start at Age
4 9 9
2 67 5 4 8 0 0
-0.40 -0.20 0.00 0.20Correlation Coefficient0.40 0.60 0.80 1.0
Males & Females Males Females
Continued Figure 5c. Studies of Tracking: HDL-C Levels Based on Age at Start of Study
Key Study, year Age at follow-up Study, year Age at follow-up (years) (years) 1. Mohler, 1996 10 20. Lauer, 1989 20-25 2. Mohler, 1996 14 21. Lauer, 1989 26-30 3. Mohler, 1996 14 22. Lauer, 1989 20-25 4. Kallio, 1993 5 23. Lauer, 1989 26-30 5. Kallio, 1993 5 24. Twisk, 1997 27-30 6. Kallio, 1993 5 25. Baumgartner, 1991 < 21 7. Kallio, 1993 5 26. Kelder, 2002 14.1 8. Bastida, 2002 4 27. Kelder, 2002 11.1 9. Porkka, 1994 15 28. Kelder, 2002 14.1 10. Porkka, 1994 18 29. Eisenmann, 2004 15.8 11. Porkka, 1994 21 30. Berenson, 1979 7 12. Porkka, 1994 24 31. Berenson, 1979 10 13. Porkka, 1994 27 32. Berenson, 1979 13 14. Porkka, 1994 30 33. Berenson, 1979 16 15. Lauer, 1989 20-25 16. Lauer, 1989 20-25 17. Lauer, 1989 20-25 18. Lauer, 1989 20-25 19. Lauer, 1989 26-30 Figure 5d. Studies of Tracking: Triglyceride Levels Based on Age at Start of Study
18 14 14
16 16 13, 15 14 4 13, 15 s 12 12 12 3 10 11 11 8 2
6 10 10
Age at Start of Study (year 1 4 9 9
2 7 6 8 0.00 0.10 0.205 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 0 Correlation Coefficient
Males & Females Males Females
Continued Figure 5d. Studies of Tracking: Triglyceride Levels Based on Age at Start of Study
Key Study, year Age at follow-up (years) 1. Berenson, 1979 7 2. Berenson, 1979 10 3. Berenson, 1979 13 4. Berenson, 1979 16 5. Kallio, 1993 5 6. Kallio, 1993 5 7. Kallio, 1993 5 8. Kallio, 1993 5 9. Porkka, 1994 15 10. Porkka, 1994 18 11. Porkka, 1994 21 12. Porkka, 1994 24 13. Porkka, 1994 27 14. Porkka, 1994 30 15. Baumgartner, 1991 < 21 16. Eisenmann, 2004 15.8 Figure 6a. Studies of Tracking: Total Cholesterol Levels Based on Duration of Study
16 38 38
14 29, 21 34, 31-36 29 30 3735 37 36 12 24 22 30 25 25 19 28 27 23, 24 23, s 44 28 26 26 10 8 22 18
8 17
5, 6 16 41 9 6710 12 13 14 Duration of Study (year 15 4 11 20 43 40 1 324 40 2 42
39 39 00.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Correlation Coefficient
Males & Females Males Females
Continued Figure 6a. Studies of Tracking: Total Cholesterol Levels Based on Duration of Study
Key Study, year Age at start Age at f/u Study, year Age at start Age at f/u (years) (years) (years) (years) 1. Berenson, 1979 5 7 24. Porkka, 1994 6 18 2. Berenson, 1979 8 10 25. Porkka, 1994 9 21 3. Berenson, 1979 11 13 26. Porkka, 1994 12 24 4. Berenson, 1979 14 16 27. Porkka, 1994 15 27 5. Porkka, 1991 3-18 9-24 28. Porkka, 1994 18 30 6. Mohler, 1996 5 10 29. Lauer, 1989 7-8 20-25 7. Mohler, 1996 10 14 30. Lauer, 1989 9-10 20-25 8. Mohler, 1996 5 14 31. Lauer, 1989 11-12 20-25 9. Kallio, 1993 0 5 32. Lauer, 1989 13-14 20-25 10. Kallio, 1993 0.17 5 33. Lauer, 1989 13-14 26-30 11. Kallio, 1993 0.33 5 34. Lauer, 1989 15-16 20-25 12. Kallio, 1993 0.50 5 35. Lauer, 1989 15-16 26-30 13. Kallio, 1993 0.75 5 36. Lauer, 1989 17-18 20-25 14. Kallio, 1993 1 5 37. Lauer, 1989 17-18 26-30 15. Guo, 1993 9-21 13-25 38. Stuhldreher, 1991 11-14 27-30 (mean age 28) 16. Guo, 1993 9-21 15-27 39. Namboodiri, 1984 <20 <20 17. Guo, 1993 9-21 17-29 40. Namboodiri, 1984 <20 <20 18. Guo, 1993 9-21 19-31 41. Kelder, 2002 8.8 14.1 19. Guo, 1993 9-21 21-33 42. Kelder, 2002 8.8 11.1 20. Bastida, 2002 0 4 43. Kelder, 2002 11.1 14.1 21. Twisk, 1997 13-16 27-30 44. Eisenmann, 2004 12-18 26.6 22. Porkka, 1994 3-18 15-30 23. Porkka, 1994 3 15 Figure 6b. Studies of Tracking: LDL-C Levels Based on Duration of Study
29, 26, 14 27, 30, 27, 28, 31, 32, 31, 19, 34 29 32 28 33 34 33 26 20, 12 21 24 24 22 25 21 25 19, 23 2023 17 22 10 8 16 s
8 15
6 5 14 11 9 6, 10 7 12
Duration of Study (year 4 18 13
35 35 32 2 1 4
00.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Correlation Coefficient
Males & Females Males Females
Continued Figure 6b. Studies of Tracking: LDL-C Levels Based on Duration of Study
Key
Study, year Age at start Age at f/u Study, year Age at start Age at f/u (years) (years) (years) (years) 1. Berenson, 1979 5 7 24. Porkka, 1994 15 27 2. Berenson, 1979 8 10 25. Porkka, 1994 18 30 3. Berenson, 1979 11 13 26. Lauer, 1989 7-8 20-25 4. Berenson, 1979 14 16 27. Lauer, 1989 9-10 20-25 5. Porkka, 1991 3-18 9-24 28. Lauer, 1989 11-12 20-25 6. Mohler, 1996 5 10 29. Lauer, 1989 13-14 20-25 7. Mohler, 1996 10 14 30. Lauer, 1989 13-14 26-30 8. Mohler, 1996 5 14 31. Lauer, 1989 15-16 20-25 9. Kallio, 1993 0.17 5 32. Lauer, 1989 15-16 26-30 10. Kallio, 1993 0.5 5 33. Lauer, 1989 17-18 20-25 11. Kallio, 1993 0.75 5 34. Lauer, 1989 17-18 26-30 12. Kallio, 1993 1 5 35. Namboodiri, 1984 <20 <20 13. Guo, 1993 9-21 13-25 14. Guo, 1993 9-21 15-27 15. Guo, 1993 9-21 17-29 16. Guo, 1993 9-21 19-31 17. Guo, 1993 9-21 21-33 18. Bastida, 2002 0 4 19. Porkka, 1994 9 21 20. Porkka, 1994 3 15 21. Porkka, 1994 6 18 22. Porkka, 1994 3-18 15-30 23. Porkka, 1994 12 24 Figure 6c. Studies of Tracking: HDL-C Levels Based on Duration of Study
31 31 28, 22, 30 22 29 24-26 23 29 27 17 15 23 30 26 25, 26 13 21 18 16 19 14 24 28 17 20 16 21, 37 19 32 12 412 s 15
10 11
8 10 1 34 3, 6 58 7 6 9 2 14 Duration of Study (year Study of Duration 4 36 33 33 38 40 41 35 2 39
0
-0.40 -0.20 0.00 0.20 0.40 0.60 0.80 1.00 Males & Females Males Females Correlation Coefficient
Continued Figure 6c. Studies of Tracking: HDL-C Levels Based on Duration of Study
Key Study, year Age at start Age at f/u Study, year Age at start Age at f/u (years) (years) (years) (years) 1. Porkka, 1991 3-18 9-24 23. Lauer, 1989 7-8 20-25 2. Mohler, 1996 5 10 24. Lauer, 1989 11-12 20-25 3. Mohler, 1996 10 14 25. Lauer, 1989 13-14 20-25 4. Mohler, 1996 5 14 26. Lauer, 1989 15-16 20-25 5. Kallio, 1993 0.17 5 27. Lauer, 1989 13-14 26-30 6. Kallio, 1993 0.50 5 28. Lauer, 1989 15-16 26-30 7. Kallio, 1993 0.75 5 29. Lauer, 1989 17-18 20-25 8. Kallio, 1993 1 5 30. Lauer, 1989 17-18 26-30 9. Guo, 1993 9-21 13-25 31. Twisk, 1997 13-16 27-30 10. Guo, 1993 9-21 15-27 32. Katzmarzyk, 2001 8-18 20-30 11. Guo, 1993 9-21 17-29 33. Namboodiri, 1984 <20 <20 12. Guo, 1993 9-21 19-31 34. Kelder, 2002 8.8 14.1 13. Guo, 1993 9-21 21-33 35. Kelder, 2002 8.8 11.1 14. Bastida, 2002 0 4 36. Kelder, 2002 11.1 14.1 15. Porkka, 1994 3-18 15-30 37. Eisenmann, 2004 12-18 26.6 16. Porkka, 1994 3 15 38. Berenson, 1979 5 7 17. Porkka, 1994 6 18 39. Berenson, 1979 8 10 18. Porkka, 1994 9 21 40. Berenson, 1979 11 13 19. Porkka, 1994 12 24 41. Berenson, 1979 14 16 20. Porkka, 1994 15 27 21. Porkka, 1994 18 30 22. Lauer, 1989 9-10 20-25 Figure 6d. Studies of Tracking: Triglyceride Levels Based on Duration of Study
12 12 14 11 14 16 15 15 13 12 10 17 13 10, 11 20 16, 10 17
8 s
6 5
67 9 8
4 Duration ofDuration (year Study
19 19 1 4 2 3 2
18 18 00.00 0.10 0.20 0.30 0.40 0.50 0.60 0.70 0.80 0.90 1.00 Correlation Coefficient
Males & Females Males Females
Continued Figure 6d. Studies of Tracking: Triglyceride Levels Based on Duration of Study
Key Study, year Age at start Age at f/u (years) (years) 1. Berenson, 1979 5 7 2. Berenson, 1979 8 10 3. Berenson, 1979 11 13 4. Berenson, 1979 14 16 5. Porkka, 1991 3-18 9-24 6. Kallio, 1993 0.17 5 7. Kallio, 1993 0.50 5 8. Kallio, 1993 0.75 5 9. Kallio, 1993 1 5 10. Porkka, 1994 3-18 15-30 11. Porkka, 1994 3 15 12. Porkka, 1994 6 18 13. Porkka, 1994 9 21 14. Porkka, 1994 12 24 15. Porkka, 1994 15 27 16. Porkka, 1994 18 30 17. Katzmarzyk, 2001 8-18 20-30 18. Namboodiri, 1984 <20 <20 19. Namboodiri, 1984 <20 <20 20. Eisenmann, 2004 12-18 26.6 Figure 7. Meta-Analysis of Statin Trials in Children With Familial Hypercholesterolemia: Total Cholesterol Reduction
Atorvastatin McCrindle, 2003 (10 mg)
Lovastatin Clauss, 2005 (20-40 mg) Stein, 1999 (10-40 mg)
Pravastatin Knipsheer, 1996 (5-20 mg) Wiegman, 2004 (20-40 mg)
Simvastatin Couture, 1998 (20 mg) De Jongh, 2002a (10-40 mg) De Jongh, 2002b (10-40 mg)
Good and Fair Trials Pooled (excluding De Jongh, 2002a) % change = 23.8 (18.6, 28.9)
All Trials Pooled % change = 24.4 (19.5, 29.2)
10 20 30 40 50 60 Mean percent change for treatment vs. control (95% Confidence Interval) Figure 8. Meta-Analysis of Statin Trials in Children With Familial Hypercholesterolemia: LDL-C Reduction
Atorvastatin McCrindle, 2003 (10 mg)
Lovastatin Clauss, 2005 (20-40 mg) Stein, 1999 (10-40 mg)
Pravastatin Knipsheer, 1996 (5-20 mg) Wiegman, 2004 (20-40 mg)
Simvastatin Couture, 1998 (20 mg) De Jongh, 2002a (10-40 mg) De Jongh, 2002b (10-40 mg)
Fair and Good Trials Pooled (excluding De Jongh, 2002a) % change = 30.2 (23.2, 37.3)
All Trials Pooled % change = 30.8 (24.1, 37.5)
10 20 30 40 50 60 Mean percent change for treatment vs. control (95% Confidence Interval) Figure 9. Meta-Analysis of Statin Trials in Children With Familial Hypercholesterolemia: HDL-C Increase
Atorvastatin McCrindle, 2003 (10 mg)
Lovastatin Clauss, 2005 (20-40 mg) Stein, 1999 (10-40 mg)
Pravastatin Knipsheer, 1996 (5-20 mg) Wiegman, 2004 (20-40 mg)
Simvastatin De Jongh, 2002a (10-40 mg) De Jongh, 2002b (10-40 mg)
Good Trials Pooled (excluding De Jongh, 2002a) % change = 3.2 (0.8, 5.7)
All Trials Pooled % change = 3.3 (1.0, 5.7)
-10 0 10 20 30 Mean percent change for treatment vs. control (95% Confidence Interval) Figure 10. Meta-Analysis of Exercise Trials in Children: Total Cholesterol Reduction
Boreham, 2000 Ferguson, 1999 Kang, 2002 Linder, 1983
Fair Trials Pooled (Ferguson, Linder) % change = -1.7 (-4.6, 7.9)
Trials of Children Pooled (Ferguson, Kang and Linder) % change = -0.6 (-6.4, 5.2)
All Trials Pooled % change = -0.0 (-5.6, 5.6)
-20 -10 0 10 20 30 Mean percent change for treatment vs. control (95% Confidence Interval) Figure 11. Meta-Analysis of Exercise Trials in Children: LDL-C Reduction
Ferguson, 1999
Kang, 2002
Linder, 1983
Savage, 1986
Fair Trials Pooled (Ferguson, Linder and Savage) % change = -5.0 (-14.7, 4.7)
All Trials Pooled % change = -3.1 (-7.7, 1.5)
-40 -30 -20 -10 0 10 20 30 Mean percent change for treatment vs. control (95% Confidence Interval) Figure 12. Meta-Analysis of Exercise Trials in Children: HDL-C Increase
Boreham, 2000 Ferguson, 1999 Kang, 2002 Linder, 1983 Stergioulas, 1998
Fair Trials Pooled (Ferguson, Linder) % change = 2.5 (-6.0, 11.1)
Trials of Children Pooled (Ferguson, Kang, Linder and Stergioulas) % change = 3.7 (-3.5, 10.8)
All Trials Pooled % change = 6.2 (-1.1, 13.6)
-25 -15 -5 5 15 25 35 45 55 Mean percent change for treatment vs. control (95% Confidence Interval) Table 1. Descriptions of Diets in Studies Name Patients Description American Heart Healthy Total fat: average of < 30% of total calories Association endorsed Saturated fatty acids: < 10% of total calories Step I Diet created by Polyunsaturated fatty acids: < 10% of total calories National Cholesterol Monounsaturated fatty acids: remaining total fat calories Education Program*36 Cholesterol: < 300 mg/day Carbohydrates: about 55% of total calories Protein: about 15-20% of total calories Calories: to promote normal growth and development and to reach or maintain desirable body weight Intended as the starting point for patients who have high cholesterol levels. Designed to reduce risk of cardiovascular disease by reducing high blood cholesterol levels.
American Heart High risk Total fat: same as Step I Association endorsed Saturated fatty acids: < 7% of total calories Step II Diet created by Polyunsaturated fatty acids: same as Step I National Cholesterol Monounsaturated fatty acids: same as Step I Education Program*36 Cholesterol: < 200 mg/day Carbohydrates: same as Step I Protein: same as Step I Calories: same as Step I Intended for people already at the Step I goals or for patients with high-risk cholesterol levels (TC> 240 mg/dL) or who had have had an MI. Designed to reduce risk of cardiovascular disease by reducing high cholesterol levels.
American Heart High risk Diet based on AHA's 1973 statement on diet and coronary heart disease for adults. Should produce comparable Association Prudent effects in children, although more research on safety is required. diet/1978 Committee Cholesterol: < 300 mg/day Report239 Calories from fat: < 35% of total calories Saturated fat: < 10% of total calories
If there is no response to the above diet, the following guidelines are recommended. For children with familial hypercholesterolemia, ages 1-10: Fat: 25-30% of total calories Substitute polyunsaturates for saturates to reach a P/S ratio of 1:1 Cholesterol: < 200 mg/day adjusted downward for younger children Familial triglyceridemia: Moderate reduction in body weight in relation to the growth curve Further reduction from 1973 statement of carbohydrates and saturated fat intake Hypercholesterolemia and Hypertriglyceridemia: Use above familial dietary restrictions. For obesity, calories must also be restricted.
1 Table 1. Descriptions of Diets in Studies Name Patients Description Dietary Intervention High risk Total fat: 28% of total energy Study in Children Saturated fat: < 8% of total energy (DISC) Diet219 Polyunsaturated fat: > 9% of total energy Cholesterol: < 75mg/day (not to exceed 150 mg/day)
Family oriented, based on social learning and social action theory.
Parent-child High risk A 10-week educational program for 4-10 year old hypercholesterolic children and their families that incorporates a Autotutorial Program hands-on and empowerment approach to learning about heart healthy food choices. Includes audio tapes that (PCAT)220 correspond with picture books for children, recommended methods to assist parents in implementing changes, and hands-on activities for both.
Key *Since 2000, the American Heart Association no longer uses the terms “Step I” and “Step II” in reference to heart-healthy diets, and have since released new guidelines.
Abbreviations P/S = Polyunsaturated/saturated, TC = Total cholesterol.
2 Table 2. Lipid Research Clinics Lipid Levels Iin U.S. Children and Adolescents
Total Cholesterol (mg/dL)*,†
Percentiles Age (years) N Mean 5 10 25 50 75 90 95 Males 0-4 238 159 117 129 141 156 176 192 209 5-9 1253 165 125 134 147 164 180 197 209 10-14 2278 162 123 131 144 160 178 196 208 15-19 1980 154 116 124 136 150 170 188 203
Females 0-4 186 161 115 124 143 161 177 195 206 5-9 1118 169 130 138 150 168 184 201 211 10-14 2087 164 128 135 148 163 179 196 207 15-19 2079 162 124 131 144 160 177 197 209
Low-Density Lipoprotein (LDL) Cholesterol (mg/dL)*,†
Percentiles Age (years) N Mean 5 10 25 50 75 90 95
White Males 5-9 131 95 65 71 82 93 106 121 133 10-14 284 99 66 74 83 97 112 126 136 15-19 298 97 64 70 82 96 112 127 134
White Females 5-9 114 103 70 75 91 101 118 129 144 10-14 244 100 70 75 83 97 113 130 140 15-19 294 99 61 67 80 96 114 133 141
High-Density Lipoprotein (HDL) Cholesterol (mg/dL)*,†
Percentiles Age (years) N Mean 5 10 25 50 75 90 95
White Males 5-9 142 57 39 43 50 56 65 72 76 10-14 296 57 38 41 47 57 63 73 76 15-19 299 48 31 35 40 47 54 61 65
White Females 5-9 124 55 37 39 48 54 63 69 75 10-14 247 54 38 41 46 54 60 66 72 15-19 295 54 36 39 44 53 63 70 76
1 Table 2. Lipid Research Clinics Lipid Levels Iin U.S. Children and Adolescents
Triglycerides (mg/dL)*,†
Percentiles Age (years) N Mean 5 10 25 50 75 90 95
Males 0-4 238 58 30 34 41 53 69 87 102 5-9 1253 30 31 34 41 53 67 88 104 10-14 2278 68 33 38 46 61 80 105 129 15-19 1980 80 38 44 56 71 94 124 152
Females 0-4 186 66 35 39 46 61 79 99 115 5-9 118 30 33 37 45 57 73 93 108 10-14 2087 78 38 45 56 72 93 117 135 15-19 2079 78 40 45 55 70 90 117 136
Key * Adapted from National Heart, Lung, and Blood Institute 198014 † All values have been converted from plasma to serum. Plasma value X 1.03 = serum value.
2 Table 3. Total Cholesterol Levels by Race*
Percentiles Race/ Ethnicity N Mean mg/dL + SEM† 50th 75th 90th 95th White 2724 162.6 + 0.537 161 180 198 209 Black 1383 172.9 + 0.830 171 192 210 225 Asian 177 165.4 + 2.120 165 181 204 219 Hispanic 1453 167.9 + 0.734 166 184 203 216 Total ‡ 5871 166.4 + 0.357 165 184 202 215
Key * Adapted from Resnicow, 198955 † 95% confidence interval = + 1.96 (standard error of the mean). ‡ Includes unclassified children
1 Table 4a. Guideline Definitions of High Risk
AAP* 1998 AHA† 2003 NCEP‡ 1992
Guidelines for High Risk LDL > 190 after 1 year low fat and low cholesterol diet X
LDL > 160 with familial premature heart disease X (before 55 years of age)
LDL > 160 with > 2 other CVD risk factors in child or adolescent X (such as low HDL cholesterol [<35 mg/dL], cigarette smoking, high blood pressure, obesity, or diabetes, etc.)
LDL > 110 mg/dL, or TC > 170 mg/dL (borderline risk), TC > 200 X mg/dL, or LDL > 130 mg/dL, or TG > 150 mg/dL, or HDL < 35 mg/dL (elevated risk)
Screen if: X Parents or grandparents with premature (< 55 years old) coronary atherosclerosis or documented MI, angina, peripheral vascular disease, cerebrovascular disease, or sudden cardiac death, or parent with total cholesterol > 240 mg/dL Risk if above is true and: LDL 110-129 mg/dL, or TC 170-199 mg/dL (borderline risk) LDL > 130 mg/dL, or TC > 200 mg/dL (high risk)
Key *American Academy of Pediatrics, 199835 †American Heart Association36 ‡National Cholesterol Education Program, 199234 Abbreviations CVD=Cardiovascular disease, HDL=High-density lipoprotein, LDL=Low-density lipoprotein, MI=Myocardial infarction, TC=Total cholesterol, TG=Triglycerides
1 Table 4b. Inclusion Criteria for Studies of Children with Familial Hypercholesterolemia
Treatment trials 71 69 68 75 74 76 80 70 81 67 83 72 77 66 82 38 79 38 73 78 Wheeler 1985 Wiegman 2004 Clauss 2005 Couture 1998 Davidson 1996 de Jongh 2002a de Jongh 2002b de Jongh 2003 Gylling 1995 Knipscheer 1996 Lambert 1996 Malloy 1978 McCrindle 1997 McCrindle 2002 McCrindle 2003 Stallings 1993 Stein 1999 Sveger 2000 Tonstad 1996a Tonstad 1996b Inclusion Criteria LDL values with or without other risk factors LDL 189 - 503 mg/dL and at least 1 parent with LDL X > 189 mg/dL not due to secondary causes LDL 220 - 503 mg/dL and a parent death from CAD X with no available lipid values LDL > 190 mg/dL with TG < 400 mg/dL X
LDL > 190 mg/dL and parent with X hypercholesterolemia (TC >290 mg/dL) or early CHD death LDL 158.5 - 398.3 mg/dL X
Two fasting LDL levels > 155mg/dL and X triglycerides < 350 mg/dL, and 1 parent with definite FH (clinical or molecular diagnosis)
LDL 160 - 400 mg/dL, and parental history of X heterozygous familial hypercholesterolemia
Fasting LDL cholesterol level before enrollment > XX 160 mg/dL and family history of hypercholesterolemia or premature CHD in first degree relatives, and minimum
1 Table 4b. Inclusion Criteria for Studies of Children with Familial Hypercholesterolemia
Treatment trials 71 69 68 75 74 76 80 70 81 67 83 72 77 66 82 38 79 38 73 78 Wheeler 1985 Wiegman 2004 Clauss 2005 Couture 1998 Davidson 1996 de Jongh 2002a de Jongh 2002b de Jongh 2003 Gylling 1995 Knipscheer 1996 Lambert 1996 Malloy 1978 McCrindle 1997 McCrindle 2002 McCrindle 2003 Stallings 1993 Stein 1999 Sveger 2000 Tonstad 1996a Tonstad 1996b Inclusion Criteria (continued) LDL values with or without other risk factors (continued) Fasting serum LDL of > 130 mg/dL while on X American Heart Association step 2 diet LDL > 95th percentile for age/gender X X
LDL > 95th percentile for age/gender or age while XXX on lipid lowering diet
LDL > 95th percentile for age/gender or age while XXX on lipid lowering diet, with documented family history of hyperlipidemia or hypercholesterolemia in siblings/parents/grandparents
LDL > 95th percentile for age while on lipid lowering XX diet, with documented family history of premature CHD < age 50 in first or second degree relatives
LDL > 95th percentile for age, while on lipid lowering X diet, and history f unsuccessul treatment with BABR.
LDL > 90th percentile for age/gender, with X triglycerides < 300 mg/dL
2 Table 4b. Inclusion Criteria for Studies of Children with Familial Hypercholesterolemia
Treatment trials 71 69 68 75 74 76 80 70 81 67 83 72 77 66 82 38 79 38 73 78 Wheeler 1985 Wiegman 2004 Clauss 2005 Couture 1998 Davidson 1996 de Jongh 2002a de Jongh 2002b de Jongh 2003 Gylling 1995 Knipscheer 1996 Lambert 1996 Malloy 1978 McCrindle 1997 McCrindle 2002 McCrindle 2003 Stallings 1993 Stein 1999 Sveger 2000 Tonstad 1996a Tonstad 1996b Inclusion Criteria (continued) LDL values with or without other risk factors (continued) Plasma LDL between 90th - 99th percentile X
TC > 300 mg/dL while on > 4 months of lipid X lowering diet
Genetic mutations Diagnosis of FH established in children and in 1 X parent mostly by DNA technique
Personal diagnosis by detection of mutation at the XXX X X LDL receptor gene
Other inclusion criteria including combinations TC > 301.6 mg/mL and tendon xanthoma in one or X both parents and in relatives in a manner compatible with autosomal dominant inheritance
TC > 290 mg/dL and first degree relative with MI X 3 Table 4b. Inclusion Criteria for Studies of Children with Familial Hypercholesterolemia Treatment trials 71 69 68 75 74 76 80 70 81 67 83 72 77 66 82 38 79 38 73 78 Wheeler 1985 Wiegman 2004 Clauss 2005 Couture 1998 Davidson 1996 de Jongh 2002a de Jongh 2002b de Jongh 2003 Gylling 1995 Knipscheer 1996 Lambert 1996 Malloy 1978 McCrindle 1997 McCrindle 2002 McCrindle 2003 Stallings 1993 Stein 1999 Sveger 2000 Tonstad 1996a Tonstad 1996b Inclusion Criteria (continued) Other inclusion criteria including combinations (continued) TC > 260 mg/dL and triglyceride levels < 200 dL, X and 1 parent had baseline cholesterol > 300 mg/dL, and triglyceride levels < 200 or tendon xanthoma, and autosomal dominant inheritance present in other members TC > 259 mg/dL, a type IIa pattern on lipoprotein X electrophoresis, and fasting triglyceride < 58.0 mg/dL in the patient, with either similar lipoprotein abnormalities in one of the parents, or premature CHD death in a parent and a similar lipid abnormality in another close relative Abbreviations BABR=Bile acid binding resin, CAD=Coronary artery disease, CHD=Coronary heart diease, CVD=Cardiovascular disease, FCH=Familial combined hyperlipidemia, FH or heFH=Familial hyperlipidemia, LDL=Low density lipoprotein, MI=Myocardial infarction, TC=Total cholesterol, TG=Triglycerides 4 Table 5a. Cholesterol Tracking by Risk Group at Baseline Age at End Length of of Study Follow-Up Study, n, and Cohort Came (years) (years) Risk Group at Baseline Results at Follow-Up for Risk Group Kallio, 1993105 5 4 TC in top quartile 40% remained in top quartile n=162 Helsinki, Finland Kelder, 20023 13 (8th 4 TC in top quintile 55% remained in top (5th quintile); 80% n=3,659 grade) remained in either 4th or 5th quintile Child and Adolescent Trial for HDL in top quintile 55% remained in top (5th) quintile; 79% Cardiovascular Health (CATCH) remianed in either the 4th or 5th quintile Laskarzewski, 197998 11-18 4 TC in top decile (13 of 108) 6 of 13 remained in top decile n=108 11-18 3 LDL in top decile (at 1 yr; 11 of 108) 3 of 11 remained in top decile Cincinnati LRC Princeton School Clarke, 197894 11-25 6 TC in lowest quintile 12% were above the 3rd quintile n=820 11-25 6 TG in middle quintile Equally distributed (20% in each quintile) Muscatine Guo, 199399 19-21 10-12 LDL>2.84 mmol/L 10% probability of LDL>3.62 mmol/L n=96 19-21 10-12 LDL>3.1 mmol/L 17% probability of LDL >3.62 mmol/L Fels Longitudinal Study 19-21 10-12 LDL>3.36 mmol/L (130mg/dL) 26% probability of LDL>3.36 mmol/L 19-21 10-12 TC>4.8 mmol/L 5% probability of TC>5.69 mmol/L 19-21 10-12 TC>5.2 mmol/L (200 mg/dL) 19% probability of TC> 5.69 mmol/L Nicklas, 2002111 and Bao, 199692 20-29 15 TC or LDL in top quintile >40% remained in top quintile n=1,169 Bogalusa Heart Study Lauer, 198995 20-30 10-14 TC>90th percentile at baseline 24-32% had adult TC>90th percentile n=2,446 20-30 10-14 Any TC measurement >90th percentile 42% had adult TC>90th percentile Muscatine during the course of study Abbreviations HDL=High-density lipoprotein, LDL=Low-density lipoprotein, TC=Total cholesterol, TG=Triglycerides 1 Table 5b. Cholesterol Tracking by Risk Group at Follow-up Length of Age at End of Follow-Up Study Study (years) (years) Risk Group at Follow-Up Derivation of the Risk Group Kallio, 1993105 5 4 TC>90th percentile 45% had TC>90th percentile at age 12 months, n=162 80% had TC>75th percentile Helsinki, Finland Porkka, 1994108 15-30 12 Highest quintile of TC Approximately 3% from 1st quintile, 11% from n=883 2nd, 12% from 3rd and 28% from 4th, 45% from Cardiovascular Risk in Young Finns 5th (highest) Study Orchard, 1983112 20-24 9-10 Adult TC >80th percentile 50% were in top quintile at baseline n= 561 20-24 9-10 All adults with TC in top 4% from 1st quintile, 11% from 2nd, 20% from Beaver County Lipid Study quintile 3rd, 18% from 4th, 48% from 5th (highest) Stuhldreher, 199196 17-30 16 Top quintile of TC 6% from 1st (lowest) quintile, 13% from 2nd, 19% n=295 from 3rd, 21% from 4th, and 40% from 5th Beaver County Lipid Study (highest) Abbreviations TC=Total cholesterol 1 Table 6. Summary of Studies Evaluating Sensitivity and Specificity of Family History Number Eligible for Number Screening Missed (based on (based on population population Study, year Population Method Threshold* Sensitivity Specificity of 1,000)† of 1,000)† Bell, 1990115 1,140 5th graders Family history of high cholesterol or MI non-fasting TC 64% 47% 540 46 1,140 5th graders Above, plus family history of stroke, non-fasting TC 77% 24% 760 31 angina or hypertension > 200 mg/dL Davidson, 1,118 4th graders Family history from parents (regarding TC > 200 41% 68% 330 83 1991116 parents, siblings, grandparents, aunts, mg/dL uncles); early MI defined as < age 56 for men and women (UCI criteria) 1,118 4th graders Parental questionnaire, definition using TC > 200 31% 66% 330 96 AAP criteria for early CHD (< age 50 mg/dL for men, < age 60 for women) Dennison, 1,214, ages 4-10, Parental questionnaire asking parental fasting TC > 38% for 73% for n/a n/a 1989126 Bogalusa Heart history of any vascular disease (CHD, 95th percentile Whites; 27% Whites; 65% Study HTN, diabetes, stroke) for AA-Blacks for AA-Blacks 2,099, ages 11-17, Parental questionnaire asking parental fasting TC > 59% for 67% for n/a n/a Bogalusa Heart history of any vascular disease (CHD, 95th percentile Whites; 25% Whites; 56% Study HTN, diabetes, stroke) for AA-Blacks for AA-Blacks 1,214,ages 4-10, Parental questionnaire asking parental fasting LDL > 41% for 73% for n/a n/a Bogalusa Heart history of any vascular disease (CHD, 95th percentile Whites; 20% Whites; 63% Study HTN, diabetes, stroke) for AA-Blacks for AA-Blacks 1 Table 6. Summary of Studies Evaluating Sensitivity and Specificity of Family History Number Eligible for Number Screening Missed (based on (based on population population Study, year Population Method Threshold* Sensitivity Specificity of 1,000)† of 1,000)† 2,099, ages 11-17, Parental questionnaire asking parental fasting LDL > 37% for 67% for n/a n/a Bogalusa Heart history of any vascular disease (CHD, 95th percentile Whites; 22% Whites; 56% Study HTN, diabetes, stroke) for AA-Blacks for AA-Blacks Diller, 232, ages 2-19, Parental questionnaire using NCEP LDL > 130 17% 75% 246 207 1995117 Cincinnati MI definition of family history of premature mg/dL Hormone Study CVD 232, ages 2-19, Parental questionnaire asking family LDL > 130 61% 74% 293 99 Cincinnati MI history of cholesterol > 240 mg/dL Hormone Study 232, ages 2-19, Both family history of elevated LDL > 130 74% 55% 478 65 Cincinnati MI cholesterol and premature CVD mg/dL Hormone Study 232, ages 2-19, Other indicators: obesity, smoking, use LDL > 130 17.4% for 86% for 547 86 Cincinnati MI of lipid raising medications, high fat mg/dL obesity, 9-48% oebsity, 69- Hormone Study diet, HTN for others 95% for others 232, ages 2-19, Family history of premature CHD LDL > 130 96% 28% 746 13 Cincinnati MI (NCEP definition), TC>240 mg/dL, or mg/dL Hormone Study any other risk factor (obesity, smoking, lipid raising medication, high fat diet or HTN). Gagliano, 224, ages 11-20 Family history of early MI ( 2 Table 6. Summary of Studies Evaluating Sensitivity and Specificity of Family History Number Eligible for Number Screening Missed (based on (based on population population Study, year Population Method Threshold* Sensitivity Specificity of 1,000)† of 1,000)† 224, ages 11-20 Family history as above, history TC above the 65% 46% 589 54 obtained from parent 85th percentile for gender Gagliano, 224, ages 11-20 Use of combined family history from TC above the 45% 69% 361 80 1993118 adolescent and parent 85th percentile (continued) for gender Griffin, 1,005, ages 2-13, Parent and grandparent history of fasting LDL > 46% NR n/a 147 1989119 8 office practices hypercholesterolemia or CHD < age 55 95th percentile 1,005, ages 2-13, Parent and grandparent history of any fasting LDL > 78% NR n/a 59 8 office practices risk factor or complication 95th percentile (hypercholesterolemia, diabetes, HTN, gout, obesity and atherosclerosis prior to age 55) 1,005, ages 2-13, Parent and grandparent history of fasting LDL > 51% 63% 385 48 8 office practices hypercholesterolemia or CHD < age 55 90th percentile 1,005, ages 2-13, Any history of parent or grandparent fasting LDL > 80% 37% 650 20 8 office practices with a risk factor or complication 90th percentile (hypercholesterolemia, diabetes, HTN, 38% for high gout, obesity and atherosclerosis prior cholesterol to age 55) alone 31% for obesity 18% for sudden death 17% for gout 13% for PVD 3 Table 6. Summary of Studies Evaluating Sensitivity and Specificity of Family History Number Eligible for Number Screening Missed (based on (based on population population Study, year Population Method Threshold* Sensitivity Specificity of 1,000)† of 1,000)† 1,005, ages 2-13, Overweight (weight for height > 95th fasting LDL > 57% NR n/a 42 8 office practices percentile) plus family history of early 90th percentile CHD or hypercholesterolermia Griffin, 1,005, ages 2-13, Overweight (weight for height > 95th fasting LDL > 84% 31.0% 704 16 1989119 8 office practices percentile) plus family history any risk 90th percentile (continued) factor or complication Muhonen, 599, ages 14-20, Parental history of high cholesterol Highest decile 34% 76% n/a n/a 1994120 Muscatine, IA of fasting TC 599, ages 14-20, Parental history of high cholesterol Highest decile 34% 76% n/a n/a Muscatine, IA of fasting LDL 599, ages 14-20, Parental history of high cholesterol Lowest decile of 26% 75% n/a n/a Muscatine, IA fasting HDL O'Loughlin, 2,217, ages 9, 13, Parental questionnaire asking personal fasting 33% 76% 256 44 2004127 and 16, Quebec history of 1) high cholesterol 2) meds LDL>109 mg/dL for cholesterol 3) heart attack, angina, ("borderline") 4) stroke, CVD or PVD or 5) meds for the heart; unknown family history coded as negative 2,217, ages 9, 13, Parental questionnaire asking personal fasting 41% 75% 256 12 and 16, Quebec history of 1) high cholesterol 2) meds LDL>131.5 for cholesterol 3) heart attack, angina, mg/dL, ("high") 4) stroke, CVD or PVD or 5) meds for the heart; unknown family history coded as negative 4 Table 6. Summary of Studies Evaluating Sensitivity and Specificity of Family History Number Eligible for Number Screening Missed (based on (based on population population Study, year Population Method Threshold* Sensitivity Specificity of 1,000)† of 1,000)† O'Loughlin, 2,217, ages 9, 13, Parental questionnaire asking personal fasting 42% 70% n/a 85 2004127 and 16, Quebec history of 1) high cholesterol 2) meds LDL>109 (continued) for cholesterol 3) heart attack, angina, mg/dL, 4) stroke, CVD or PVD or 5) meds for ("borderline") the heart; unknown family history excluded 2,217, ages 9, 13, Parental questionnaire asking personal fasting 51% 69% n/a 19 and 16, Quebec history of 1) high cholesterol 2) meds LDL>131.5 for cholesterol 3) heart attack, angina, mg/dL, ("high") 4) stroke, CVD or PVD or 5) meds for the heart; unknown family history excluded Primrose, 1,012, ages 12-15, History of stroke, angina or heart attack non-fasting TC 33% 72% 293 125 1994121 Ireland in either parent at any age or in 1st > 95th degree grandparents, uncles or aunts < percentile age 55. Questionnaires completed by according to parents LRC Resnicow, 574, elementary Parental cholesterol >240 in 1 parent non-fasting TC 10% 91% 90 106 1993122 school age only with known and reported value by > 200 mg/dL that parent Rifai, 199662 260, ages 12-20, Family history of early CHD or fasting LDL > 10% NR 365 184 African American- hyperlipidemia 110 mg/dL Black 5 Table 6. Summary of Studies Evaluating Sensitivity and Specificity of Family History Number Eligible for Number Screening Missed (based on (based on population population Study, year Population Method Threshold* Sensitivity Specificity of 1,000)† of 1,000)† Sanchez 2,224, ages 2-18, Parental history of MI fasting TC>200 7% 96% 49 140 Bayle, Spain mg/dL 1992123 Sanchez 2,224, ages 2-18, Parental history of MI fasting 9% 96% 49 101 Bayle, Spain LDL>135 mg/dL 1992123 (continued) 2,224, ages 2-18, Parental history of stroke, HTN, fasting TC>200 14% 90% 98 129 Spain diabetes, or hypercholesterolemia (but mg/dL not MI) 2,224, ages 2-18, Parental history of stroke, HTN, fasting 14% 91% 98 95 Spain diabetes, or hypercholesterolemia (but LDL>135 mg/dL not MI) Shea, 199059 108, ages 4-5, AAP definition (maternal hypertension, fasting TC > 57% 59% 493 148 Hispanic, Study of diabetes, obesity, hyperlipidemia or 170 mg/dL Childhood Activity family history of premature CHD or & Nutrition hyperlipidemia) 108, ages 4-5, AHA and NIH Consensus Conference fasting TC > 46% 70% 352 185 Hispanic, Study of definition (history of hyperlipidemia or 170 mg/dL Childhood Activity premature CHD in the child's parent, & Nutrition aunt, uncle or grandparent) 108, ages 4-5, NCEP guidelines (history of MI or fasting TC > 5% 92% 74 324 Hispanic, Study of sudden death in the child's parent, 170 mg/dL Childhood Activity aunt, uncle, or grandparent; CHD prior & Nutrition to age 55). 6 Table 6. Summary of Studies Evaluating Sensitivity and Specificity of Family History Number Eligible for Number Screening Missed (based on (based on population population Study, year Population Method Threshold* Sensitivity Specificity of 1,000)† of 1,000)† Steiner, 1,001, ages 12-21 AAP 1998 criteria (known non-fasting 63% 60% 400 24 1991124 (38% Hispanic, hyperlipidemia in parent or sibling, TC>200 mg/dL, 15% Black, 11% known MI/angina, current corticosteroid repeated fasting Asian, 33.5% use, juvenile diabetes, hypothyroidism, TC if initial White), Kaiser renal/endocrine/hepatic disease in test>200 mg/dL, population teenager) repeated a 3rd time if more than 30 mg/dL variability between the 1st two measurements Troxler, 110 mostly Questionnaires completed with parental fastingTC > 38% 79% 218 245 1991125 Hispanic senior assistance; family history in parents or 75th percentile high school grandparents of high cholesterol or (4.53 mmol/L) students CHD age <55 (AAP) Wadowski, 300 African Family history of CHD in parent or fasting TC > 59% 72% 327 23 1994128 American-Blacks, grandparent at age < 55 215 mg/dL ages 2-14 Key *If not explicity stated, values are mixed non-fasting/fasting or not reported. †Number eligible for screening and number missed were calculated from available data. In some cases, reported data did not allow for these calculations (these indicated with n/a). Abbreviations AA-Blacks=African-American Blacks, AAP=American Academy of Pediatrics, AHA=American Heart Asssociation, CHD=Coronary heart disease, CVD=Cardiovascular disease, HTN=Hypertension, LDL=Low-density lipoprotein, MI=Myocardial infarction, NCEP=National Cholesterol Education Program, NIH=National Institutes of Health, PVD=Peripheral vascular disease, TC=Total cholesterol 7 Table 7a. Summary of Risk Factors for Elevated Total Cholesterol in Children and Adolescents Aerobic capacity/ Overweight or other Other Physical cardiovascular Biological Composition Biological and Study Activity fitness Diet Family History Measures Miscellaneous Bonora, 1992137 Total body fat, M: 0 Bonora, 1996138 Total body fat, M: 0 Demerath, 2003139 X+ DeStefano, X+ X+ BMI: X+* Parent smokes: X-* 1995140 Douglas, 1996141 X+ X+ DuRant, 1982142 Parent stroke, diabetes, obesity: X+* Family history of diabetes, children only:O* Family history of diabetes, adolescent only: X+* DuRant, 1983143 M: X+* M: 0 General, M: 0 Smoking, M: 0 Alcohol, M: 0 Television watched per night, M: X+* DuRant, 1993144 00 Dwyer, 1994145 00 X- Eisenmann, SSF, Adolescent, M: X+* 2002146 Freedman, 199918 Overweight: X+ 1 Table 7a. Summary of Risk Factors for Elevated Total Cholesterol in Children and Adolescents Aerobic capacity/ Overweight or other Other Physical cardiovascular Biological Composition Biological and Study Activity fitness Diet Family History Measures Miscellaneous Giovannini, CVD: 0 F: X+ 1992148 HC parents: X+ Glassman, BMI, %ideal weight: X+* 1993149 Obese: 0 BMI in non-obese, HC: X- Gliksman, Dietary fat, M: 0 BMI, M: X- 1993174 Howard, 1991150 Grain: X+ Jarvisalo, 2001151 aIMT, cIMT, HC, DM: X+ cIMT: 0 Kunz, 2005153 Smoking, M: X- Larsson, 1992152 FHx hyperlipidemia: X+ FHx MI/angina exclusively: 0 Macek, 1989155 0 Marti, 1989156 00 Muhonen, 1994120 Grandfather hx, HC: X+ Resnicow, 1993122 BMI: X+ 2 Table 7a. Summary of Risk Factors for Elevated Total Cholesterol in Children and Adolescents Aerobic capacity/ Overweight or other Other Physical cardiovascular Biological Composition Biological and Study Activity fitness Diet Family History Measures Miscellaneous Ribeiro, 2004159 Body fatness: 0 Shear, 1985160 Single parental vascular disease: 0 Parent MI & parent DM or HTN: X+ Simon, 1995161 F: 0 F: 0 Suter, 1993162 BMI: 0 Thorland, 1981163 M: X+* 0 Body fatness: 0 Tolfrey, 1999164 F: X+ F: X+* P-fat, F: X+* %Body fat, F: X+ Caloric intake & M: 0 TC, M: X+ Tonstad, 1995165 Puberty stage, M: X- Twisk, 2001166 Oxygen uptake: SSF: X+* X+* Twisk, 1998167 BMI, TC: X+* SSF, TC: X+* Lean Body Mass: 0* Twisk, 1996168 P/S fat: X- Smoking: X+* Cholesterol: X+ van Lenthe, Change in S/T, M: 0* 1998169 F: 0 3 Table 7a. Summary of Risk Factors for Elevated Total Cholesterol in Children and Adolescents Aerobic capacity/ Overweight or other Other Physical cardiovascular Biological Composition Biological and Study Activity fitness Diet Family History Measures Miscellaneous van Stiphout, 0 Coffee: 0 TC of parents & TC of Age 15+: X+ Menarche: 0 1985170 children: X+ Age <15: 0 Age 9-16: X- Ward, 1980171 X+ X+ Wong, 1992172 0 Extra lean/no High TC: X+ BMI: 0 2-4 hours TV/day: X+ hamburger vs Premature MI: 0 Age: 0 lean/regular hamburger: X+ Key X+ positive significant relationship X- negative significant relationship 0 no significant relationship * TC/HDL Abbreviations aIMT=Abdominal aorta intima medial thickness, cIMT=Cartoid artery intima medial thickness, DM=Diabetes mellitus, Dyslip=Dyslipidemia, F=Female only, FHX=Family history, HC=High cholesterol, HDL=High-density lipoprotein, HTN=Hypertension, HX=History, M=Male only, MI=Myocardial infarction, P-fat=Polyunsaturated fat, P/M=Polyunsaturated/monosaturated fat, P/S=Polyunsaturated/saturated fat, S-fat=Saturated fat, SSF=Sum of skinfolds, S/T=Subscapular/tricep skinfold ratio, T/S=Tricep/suprailliac 4 Table 7b. Summary of Risk Factors For Elevated LDL in Children and Adolescents Aerobic capacity Overweight or other Other Physical /cardiovascular Biological Composition Biological and Study Activity fitness Diet Family History Measures Miscellaneous Bergstrom, 1997136 X+ Bonora, 1992137 Total body fat, M: 0 Bonora, 1996138 Total body fat, M: 0 DeStefano, 1995140 X+ X+ DuRant, 1982142 Parent stroke, DM, obesity, LDL/HDL: X+ FHX of stroke, LDL/HDL, children only: X- FHX of DM, LDL/HDL, children only: 0 LDL/HDL, adolescent only: X+ Obese family member, LDL/HDL, M: 0 LDL/HDL, F: X+ DuRant, 1983143 LDL/HDL, 0 General, M: 0 Smoking, M: 0 M: X+ Alcohol, M: 0 Hours reading/day, LDL/HDL, M: X+ DuRant, 1993144 0 Waist/hip year 1: X+ Eisenmann, 2002146 SSF, Adolescents, M: X+ Freedman, 199918 Overweight: X+ 1 Table 7b. Summary of Risk Factors For Elevated LDL in Children and Adolescents Aerobic capacity Overweight or other Other Physical /cardiovascular Biological Composition Biological and Study Activity fitness Diet Family History Measures Miscellaneous Giovannini, 1992148 HC parents: X+ F: X+ M: X- Glassman, 1993149 BMI, %ideal weight, in obese: 0* BMI, HC, in non-obese: X- Gliksman, 1993174 PS ratio, LDL/HDL, BMI, LDL/HDL: X+ F: X+ Dietary fat, LDL, LDL/HDL, M: 0 Kunz, 2005153 Smoking, M: X- Jarvisalo, 2001151 cIMT: 0 Larsson, 1992152 FHX, hyperlipidemia: X+ FHX, MI or angina exclusively: 0 Kwiterovich, 1997154 Dietary BMI at year 3: X+ Baseline LDL: X+ cholesterol, M: X+ Any single nutrient, Tanner Stage: X- F: 0 Macek, 1989155 X- Muhonen, 1994120 Grandfather hx of HC: X+ 2 Table 7b. Summary of Risk Factors For Elevated LDL in Children and Adolescents Aerobic capacity Overweight or other Other Physical /cardiovascular Biological Composition Biological and Study Activity fitness Diet Family History Measures Miscellaneous Suter, 1993162 VLDL: X+ VLDL: 0 Dietary fat, VLDL: BMI, VLDL: 0 0 Tolfrey, 1999164 F: X+ LDL/HDL, F: X+ P-fat & LDL, F: X+ %Body fat, F: X+ Caloric intake & LDL, M: X+ Tonstad, 1995165 M: X+ LDL & parent %Body fat in those with FH Puberty stage, M: cholesterol, LDL & non- mutation only: X+ X- FH parent's cholesterol: X+ Key X+, positive significant relationship X-, negative significant relationship 0, no significant relationship * TC/HDL Abbreviations BMI=Body mass index, cIMT=Cartoid artery intima medial thickness, DM=Diabetes mellitus, F=Female only, FHX=Family history, HC=High cholesterol, HDL=High-density lipoprotein, LDL=Low-density lipoprotein, M=Male only, MI, myocardial infarction, P-fat=Polyunsaturated fat, P/S=Polyunsaturated/saturated fat, SSF=Sum of skinfolds, VLDL=Very low-density Lipoprotein 3 Table 7c. Summary of Risk Factors for Decreased HDL in Children and Adolescents Aerobic capacity Overweight or other Other Physical /cardiovascular Biological Composition Biological and Study Activity fitness Diet Family history Measures Miscellaneous Andersen, 2003135 X+ X+ Bergstrom, 1997136 X+ Fat intake, M: X+ BMI: X- Bonora, 1992137 Total body fat, M: X+ Bonora, 1996138 Total body fat, M: X- Demerath, 2003139 X- DeStefano, 1995140 X+ X+ BMI: X+ Parent smokes: X-* DuRant, 1982142 Parent stroke, DM, obesity, LDL/HDL: X+* Family history of stroke, LDL/HDL, children only: X- Family history of DM, LDL/HDL, children only: 0* LDL/HDL, adolescent only: X+* FHX obesity, LDL/HDL, M: 0 LDL/HDL, F: X+ DuRant, 1983143 LDL/HDL, M: 0 General, M: 0 Smoking, M: 0 M: X+* Alcohol, M: 0 Television watched per night: X+* HDL: X- Hours reading per day, LDL/HDL, M: X+ DuRant, 1993144 SSF Year 2: X+ 1 Table 7c. Summary of Risk Factors for Decreased HDL in Children and Adolescents Aerobic capacity Overweight or other Other Physical /cardiovascular Biological Composition Biological and Study Activity fitness Diet Family history Measures Miscellaneous Dwyer, 1994145 00 BMI: X- SSF: X+ Eisenmann, SSF, adolescent, M: X+* 2002146 HDL, adolescent, F: X- Freedman, 199918 Overweight: X- Fripp, 1985147 M: X+ BMI, M: X- Giovannini, 1992148 FHX, CVD: 0 F: X- Gliksman, 1993174 % saturated fat, BMI: X- PS ratio, F: X+ M: 0 Howard, 1991150 High fat: X+ Kunz, 2005153 Smoking, M: O Jaing, 1995173 M: X+ SSF, change in Caucasians ages 5- F: 0 subscapular thickness: 17: X- X+ BMI, F: X+ Jarvisalo, 2001151 cIMT: 0 Larsson, 1992152 FHX MI/angina or angina exclusively, LDL/HDL: 0 Any family history: 0 Macek, 1989155 X+ Marti, 1989156 00 BMI: X- Sexual maturation, M: X- 2 Table 7c. Summary of Risk Factors for Decreased HDL in Children and Adolescents Aerobic capacity Overweight or other Other Physical /cardiovascular Biological Composition Biological and Study Activity fitness Diet Family history Measures Miscellaneous Muhonen, 1994120 Grandfather hx of CHD: X+ Grandfather hx of high cholesterol, LDL/HDL: X+ Raitakari, 1996157 M: X+ BMI change, F: X+ F: 0 SSF: X+ Shear, 1985160 Single parental vascular disease: 0 Parent MI & parent DM or HTN: X+ Simon, 1995161 F: 0 P/S fat, PM fat, P- SSF: X- fat: 0 T/S SF, F: X+ F: X+ Suter, 1993162 X+ 0 Dietary fat: 0 BMI: 0 SSF: X- Thorland, 1981163 M: X+* M: 0 Body fatness, M: 0 Tolfrey, 1999164 HDL,LDL/HDL, F: P-fat, F: X+* X+* Tonstad, 1995165 HDL & FH parent HDL: X+ Puberty stage, M: X- Twisk, 2001166 Oxygen uptake SSF X+* X+* Twisk, 1998167 BMI: 0 SSF: 0 Lean Body Mass: 0 Twisk, 1996168 X+ Alcohol: X+ Smoking: X- 3 Table 7c. Summary of Risk Factors for Decreased HDL in Children and Adolescents Aerobic capacity Overweight or other Other Physical /cardiovascular Biological Composition Biological and Study Activity fitness Diet Family history Measures Miscellaneous van Lenthe, 1998169 S/T, M: X- Change in S/T, F: 0 Change in TC/HDL, M: 0 van Stiphout, Age 15, M: X- 1985170 Key X+ positive significant relationship X- negative significant relationship 0 no significant relationship * TC/HDL Abbreviations BMI=Body mass index, CHD=Coronary heart disease, cIMT=Cartoid artery intima medial thickness, CVD=Cardiovascular disease, DM=Diabetes mellitus, F=Female only, FHX=Family history, HDL=High-density lipoprotein, HTN=Hypertension, LDL=Low-density lipoprotein, M=Male only, MI=Myocardial infarction, P-fat=Polyunsaturated fat, P/M=Polyunsaturated/monosaturated fat, P/S=Polyunsaturated/saturated fat, SSF=Sum of skinfolds, S/T=Subscapular/tricep skinfold ratio, T/S=Tricep/suprailliac, SF=Saturated fat 4 Table 8. Randomized Controlled Trials of Drug Treatment for Children with Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Author, Duration of Within Group Quality year Drug Population Trial Baseline Diet Between Group Differences Differences Rating Statins Clauss, Lovastatin 20 n=54, girls 24 weeks NR Total cholesterol, LDL, and Good 200538 mg/day for 1st ages 10-17, Apo B were significantly 4 weeks then all post- lower after 4 wks (on 20mg) 40 mg/day for menarchal and after 24 wks (on 40mg) remaining 20 compared to placebo. At 24 weeks vs. wks, mean % changes from placebo baseline were: TC: 26.8% for lovostatin vs +5.2 for placebo (p<0.001); LDL: - 21.8 for lovastatin vs +4.5 for placebo (p<0.001). Couture, Simvastatin 20 n=63 6 weeks AHA Phase 1 diet Total cholesterol, LDL, and Fair 199866 mg/day vs. total ApoB levels were placebo significantly lowered by treatment (vs. placebo) at all time measurements (p<0.0001). HDL levels increased in all groups with drug vs. placebo (p=0.003 to 0.30). TG levels decreased (p=0.009 to 0.10). 1 Table 8. Randomized Controlled Trials of Drug Treatment for Children with Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Author, Duration of Within Group Quality year Drug Population Trial Baseline Diet Between Group Differences Differences Rating de Jongh, Simvastatin 10 n=50, ages 9- 28 weeks NR Mean absolute change in Treatment group Poor 200268 mg/day, 18 FMD was higher for had reductions in doubled every treatment group vs. placebo mean TC 8 weeks up to (p=0.05). (p=0.0001), LDL 40 mg/day vs. TC, LDL and TG were all (p=0.0001) and TG placebo reduced for treament group (p=0.041). vs. placebo (p=0.0001, No significant p=0.0001 and p=0.04 changes in respectively). lipoproteins for placebo group compared with baseline. FMD increased in the treatment group (p<0.0001) and did not change in the placebo group. de Jongh, Simvastatin 10 n=173, ages 48 weeks of NR TC and LDL-C, ApoB, and Good 200269 mg/day titrating 10-17 intervention VDL-C reduced at all time up to 40 after 4 points by drug vs. placebo mg/day vs. week run-in (p<0.001). placebo Knipscheer Pravastatin in 3 n=72 12 weeks, Carbohydrates TC, LDL and ApoB100 were TC, LDL and Good , 199671 active drug following an 50% of energy, significantly reduced in all ApoB100 were groups: 5, 10, 8 wk diet protein 20%; and pravastatin treated groups significantly reduced or 20 mg/day and fat 30%; compared with placebo in all pravastatin vs. placebo placebo run- unsaturated: (p<0.05 for both). treated groups in period. saturated ratio of HDL increases and TG compared with 2:1; total changes were not baseline (p<0.05). cholesterol <300 significantly different from HDL increases and mg/day placebo. TG changes were not significantly different from baseline . 2 Table 8. Randomized Controlled Trials of Drug Treatment for Children with Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Author, Duration of Within Group Quality year Drug Population Trial Baseline Diet Between Group Differences Differences Rating Lambert, Lovastatin at n= 69, boys 8 weeks Patients Demonstrated a dose All doses reduced Fair 199672 10, 20, 30, or age < 17 counseled by response relationship, with total cholesterol, 40 mg/day (4 dietician improved results up to a dose LDC, and ApoB (P < active drug throughout the of 30 mg/day. .0001 for all). groups, no trial; fat 30% total placebo) calories, saturated fat <10%, polyunsaturated fats <10%, cholesterol <125 mg/100 kcal/day or <300 mg/day (whichever less). 3 Table 8. Randomized Controlled Trials of Drug Treatment for Children with Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Author, Duration of Within Group Quality year Drug Population Trial Baseline Diet Between Group Differences Differences Rating McCrindle, Atorvastatin 10 n=187, ages 26 weeks NCEP Step 1 diet TC, LDL-C, TG and ApoB Reductions in TC, Good 200376 mg/day vs. 10-17 were reduced compared to LDL and TG and placebo placebo (p<0.001 for TC, increase in HDL p<0.001 for LDL-C, p=0.03 were significant for for TG, p<0.001 for apoB). simvastatin group. HDL cholesterol increased No significant for treatment vs. placebo improvements in (p=.02). placebo group. Stein, Lovastatin 132 boys, 48 weeks AHA pediatric diet TC and LDL-C decreased at Good 199978 starting at ages 10-17 all dosages (p<.001). About 10mg/day, 6% additional LDL-C titrating to 40 lowering occurred with each mg/day vs. doubling of lovastatin placebo dosage. No significant changes in HDL-C or triglycerides compared to placebo. Wiegman, Pravastatin 40 214 ages 8- 2 years Instructed to TC decreased for pravastatin Good 200483 mg/day vs. 18 continue a fat vs. placebo (56% vs 2%, placebo restricted diet p<0.001) LDL decreased 57% for pravastatin vs 0% for placebo 9<0.001) No significant differences for HDL or placebo. Carotid IMT decreased for pravastin vs. placebo (p=0.02) 4 Table 8. Randomized Controlled Trials of Drug Treatment for Children with Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Author, Duration of Within Group Quality year Drug Population Trial Baseline Diet Between Group Differences Differences Rating Bile-acid Resins McCrindle, Cholestyramin n=40, ages 8 weeks AHA Step II diet No lipid levels reported prior Reduction in LDL for Fair 199774 e 8 gm/day as 10-18 each cross- to cross-over. both pills (p=0.006) pills vs. powder over period At end of study, 82% and powder (no placebo) preferred pill, 16% powder, (p=0.0001) with no and 2% neither form. Mean significant difference compliance was greater for between forms pills (p=0.01). (p=0.16). McCrindle, Colestipol (5 n=40, ages 8- 18 weeks AHA Step II diet No data reported prior to Reductions in TC Poor 200275 g/day) + 18 each cross- cross-over. and LDL seen with pravastatin 10 over period both colestipol only mg/day vs. and colestipol plus colestipol alone pravastatin. (10g/day) Tonstad, Colestipol 10 66 8 weeks NCEP diet: fat TC and LDL decreased for Poor 199680 gm/day or 5 adolescents <30% of calories, treatment group vs. placebo gm twice daily saturated fat (p<0.01 for both). HDL and vs. placebo <10%, cholesterol TG were not signficantly <200 mg/day. different. Tonstad, Cholestyramin 96 boys 1 year NCEP diet: fat At 1 year: Fair 199681 e titrating up ages 6-11 <30% of calories, TC decreased 11.5% in from 4 gm/day saturated fat treatment vs 3% in placebo to 8 gm/day vs. <10%, cholesterol (p<0.001) placebo <200mg/day. LDL decreased 18.6% in treatment group vs. 1.5% in the placebo group (p<0.0001). No changes in HDL or TG. 5 Table 8. Randomized Controlled Trials of Drug Treatment for Children with Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Author, Duration of Within Group Quality year Drug Population Trial Baseline Diet Between Group Differences Differences Rating Fibrate Wheeler, Bezafibrate 10- 14 3 months Low saturated fat, No data reported prior to Mean plasma TC Poor 198582 20 mg/kg/day each cross- increased cross-over. decreased during twice daily vs. over period polyunsaturated bezafirate treatment placebo fat; actual total fat (p<0.0001) intake of 98g/day Mean HDL (range: 44-148 increased (p<0.01). gm/day) No change in TG. Other Engler, Vitamin C 250 n=15, ages 8- 6 weeks Food guide No data given prior to cross- NCEP II diet Poor 2003224 mg twice daily 21 each cross- pyramid and over. resulted in an 8% vs. Vitamin E over period NCEP II reduction in LDL 200 IU twice guidelines (p<0.01) from daily vs. baseline. No other placebo significant differences in lipid levels. Brachial artery FMD improved compared with baseline (p<0.001) in response to diet plus anti-oxidants. 6 Table 8. Randomized Controlled Trials of Drug Treatment for Children with Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Author, Duration of Within Group Quality year Drug Population Trial Baseline Diet Between Group Differences Differences Rating Engler, Docosahexaen 20 children 6 weeks NCEP II No data reprted prior to cross-For combined Poor 2004234 oic acid vs. ages 9-19 each cross- over. groups, FMD placebo over period, decreased with 6 week significantly from wash-out baseline for both DHA and placebo groups. TC, LDL and HDL increased significantly from baseline for DHA group, but not for placebo group. Engler, Docosahexaen 20 children 6 weeks NCEP II No data reported prior to For combined Poor 2005234 oic acid vs. ages 9-19 each cross- cross-over. groups, there were placebo over period, no significant with 6 week differences in TC, wash-out LDL, HDL or triglycerides. The DHA +diet group had lower LDL subfractions (LDL1 and LDL 3) and higher HDL2 compared to placebo + diet. 7 Table 8. Randomized Controlled Trials of Drug Treatment for Children with Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Author, Duration of Within Group Quality year Drug Population Trial Baseline Diet Between Group Differences Differences Rating Malloy, p- 20 children 6 months Restricted in No data reported prior to Mean TC was lower Poor 1978236 Amnosalicylic ages 5-21 each cross- cholesterol (<200 cross-over. for treatment with acid 150 ver period g/day) and diet and PAS-C vs. mg/kg/day up saturated fat diet alone (p<.001). to 8 g/day vs. (<20% of total Mean fasting serum placebo calories); all triglyceride levels participants and decreased 15.7% parents received with PAS-C plus diet monthly dietary (p<.001). Levels in counseling the compliant subgroup were not lower than those of other patients. No difference in HDL with drug treatment. Abbreviations AHA=American Heart Association, ApoA=Apolipoprotein A, ApoB=Apolipoprotein B, DHA=Docosahexaenoic acid, FMD=Flow-mediated dilation, HDL=High-density lipoprotein, HDL-C=High-density lipoprotein cholesterol, IMT=Intima medial thickness, LDL=Low-density lipoprotein, LDL-C=Low-density lipoprotein cholesterol, NCEP=National Cholesterol Education Program, NR=Not reported, PAS-C=P-aminosalicylic acid-cholesterol, TC=Total cholesterol, TG=Triglycerides, VDL-C=Very low- density Lipoprotein cholesterol 8 Table 9. Randomized Controlled Trials of Diet Treatment for Children with Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Duration Between Group Within Group Quality Author, year Intervention(s) Population of Trial Comparisons Comparisons Rating Amundsen, Plant sterol spread vs. 38 children 8 weeks No data reported prior to Results combined indicate: Fair 200241 Control spread each cross-over. TC, LDL, Apo B were period decreased for treatment group vs. control. No change in HDL, TG, or Apo A-1. Davidson, 58 gram psyllium- 32 children 6 weeks No data reported prior to At 6 weeks: Fair 199667 enriched cereal vs. aged 6 - 18 each cross-over. TC decreased 8% in placebo cereal within the period of psyllium group (p=0.05) vs Chicago area cross over 3% for placebo (NS). LDL cholesterol decreased 10% in psyllium group (p=0.001) vs 0.5% in placebo group (NS). Response to treatment differed between subjects consuming psyllium during period 1 vs period 2. A period-by-time interaction was found for total cholesterol (p=0.02). 1 Table 9. Randomized Controlled Trials of Diet Treatment for Children with Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Duration Between Group Within Group Quality Author, year Intervention(s) Population of Trial Comparisons Comparisons Rating Gylling, Rapeseed oil-rich 14 children 6 weeks No data reported prior to TC decreased 11% in Poor 199539 margarine with with HeFH each arm cross-over. sitostanol group vs 0.8% in sitostanol ester vs. (mostly cross over placebo group (p<0.05). Rapeseed oil-rich established by LDL 18% decreased in margarine without DNA) 7 sitostanol group vs 3% in sitostanol ester each group placebo group (p<0.05). No change in HDL or TG. de Jongh, Sterol-enriched 41 prepubertal 4 weeks Last 4 weeks: TC and Good 200340 margarine, (15g/day) HeFH, ages 5-each LDL were decreased vs. Identical appearing 12 period significantly for sterol control spread, non- cross over group vs placebo sterol-enriched (p<0.001 for both). No (15g/day) difference in HDL, TG, or FMD for sterol vs. placebo. McCrindle, 300 mg tablet of garlic 30 children 8 weeks Relative to placebo, garlic Fair 1998235 extract (Kwai, Lichtwer with FH, ages treatment resulted in: Pharma, Berlin, 8-18 TC increase 0.6 mmol/L Germany) containing (p=0.86) 0.6 mg of allicin placed LDL-C decrease in a gelatin capsule 0.5mmol/L, p=0.90 with inert filler vs HDL increase 9.3 mmol/L, identical capsule with p=0.29 gelatin filler only. TG decrease 0.72 mmol/L, p=0.70. Abbreviations ApoA-1=Apolipoprotein A-1, ApoB=Apolipoprotein B, FH or HeFH=Familial hyperlipidemia, FMD=Flow mediated dilation, HDL=High-density lipoprotein, LDL=Low- density Lipoprotein, LDL-C=Low-density lipoprotein cholesterol, NS=Not significant, TC=Total cholesterol, TG=Triglycerides 2 Table 10. Randomized Controlled Trials of Diet Treatment for Children Without Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Author, year Intervention(s) Control Population Dennison, 1993223 Psyllium cereal (3 g of water-soluble fiber and Placebo cereal with 5 gm of water- 25 children ages 5- 3 gm of water-insoluble fiber per serving). insoluble fiber. Two 28 gm servings per 17 with LDL>110 Two 28 gm servings per day. day. mg/dL after diet treatment 12 each group DISC Collaborative Family oriented behavioral intervention to Usual care: Public educational 663 prepubertal Research Group, promote dietary adherence. Sessions weekly x publications on heart-healthy eating boys and girls, 1995219 6 weeks, then biweekly x 10 weeks, plus 2 provided. Parents informed if child blood ages 8-10 individual visits for children with family cholesterol high - no specific members. In second 6 months: 4 group and 2 recommendations to see physician individual sessions. Years 2 and 3: group given. 3-year lipid results provided, with individual and maintenance sessions held 4-6 referral as clinically warranted. x per year with monthly phone contacts as needed. Used motivational interviewing and stages of change. Gold, 1991231 Oat bran supplemented cereal within AHA Control cereal within AHA Step 1 diet 49 school age Step 1 diet children with TC>185 mg/dL Kuehl, 1993227 MSI= four 90 minute sessions with focus on SSI=one 90 minute nutrition education 295 children ages 2- food preparation. Participants received session for pt, siblings and parents 15 with TC >185 notebooks with nutrition information and (slides presentation, low fat food prep recipes, incentives for attendance and and tasting, distribution of fruit and completion of behavioral contracts (eating low cereal). "Returned for subsequent food fat meal) Multi-session intervention. sampling with nutritionist present" but no further formal education. Single session intervention. 1 Table 10. Randomized Controlled Trials of Diet Treatment for Children Without Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Quality Author, year Between Group Comparisons Within Group Comparisons Rating Dennison, 1993223 No pre cross-over results given. Poor No difference for changes in TC, HDL or LDL. TG increased less for psyllium group vs. control (p<0.05). DISC Collaborative TC decreased for diet group vs. control in year 1 Good Research Group, (p=0.03) and year 3 (p=0.04) 1995219 LDL decreased in the diet group vs. control at 1 and 3 years (p<0.001 and p=0.02). HDL decreased in diet group vs. control between groups at year 1 (p=0.03) but not year 3 (p<0.75). No significant differences between groups for TG at years 1 or 3. Dietary total fat, saturated fat, and cholesterol intakes were lower in the diet group vs. control. Gold, 1991231 No significant differences between groups in total Poor cholesterol, LDL, HDL, TG or Apo A. Intervention group had decreased ApoB (-9mg/dL) vs. control (2mg/dL), p=0.05. Kuehl, 1993227 Both groups had improvement in TC from Poor baseline to the 16-week visit (no results given for between group comparison). MSI group also had significantly decreased LDL. Both groups had a significant decreases in the proportion of total calories obtained from fat and the proportion of total calories obtained from protein. Total calorie intake decreased in the MSI group but not the SSI group. 2 Table 10. Randomized Controlled Trials of Diet Treatment for Children Without Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Author, year Intervention(s) Control Population Obarzanek, 2001218 Counseling intervention, same as DISC above. Control: Usual care - parents informed 663 randomized, child's blood cholesterol high and were 580 at last follow- given educational materials on heart- up visit healthy eating as available to the public. Shannon, 1994220 Parent-Child AutoTutorial Program (PCAT) At-risk control group. Ages 4-10 with LDL included 10 talking book lessons and follow-up 107-164 for boys paper and pencil games for children with a and 112-164 for manual for parents. Counseling treatment: 45- girls (mg/dL) 60 minute session with parent, child and registered dietitian, along with take home print materials for both. Dietitian available for phone to answer questions during 3 month period. Stallings, 199377 Parent-Child AutoTutorial Program (PCAT) 10 Usual Care: 1 hour session with 44 children ages 4- sessions total, 1 per week completed in home registered dietician using AHA guide 10 with LDL by child and parents between 90-99th percentile. 3 Table 10. Randomized Controlled Trials of Diet Treatment for Children Without Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Quality Author, year Between Group Comparisons Within Group Comparisons Rating Obarzanek, 2001218 Results for years 1 and 3 are the same as presented in Good DISC study above Results for years 5 and 7 follow-up presented here. No difference between intervention group and control for LDL at 5 years (p=0.11) or final visit (yr 7) (p=0.25). No difference for TC at 5 or 7 years. No difference for HDL at 5 or 7 years. No differences in TG or LDL/HDL ratio. Intervention group had lower dietary total fat, saturated fat at all points (p<0.001) and lower cholesterol intake at year 5 (p<0.001). Shannon, 1994220 Mean LDL-C decline of PCAT (10.1mg/dL) was greater Good than in at-risk control (4.1 mg/dL) (p<.05). This represents an 8% decline in the PCAT group. Dietary knowledge scores: increased 3x more in PCAT vs. counseling or at-risk control (p<.001). Lipid intake: Mean grams of total and saturated fat decreased in PCAT/counseling group and increased slightly in at-risk control (p<.05). Stallings, 199377 No between group differences at either 3 or 6 months LDL decreased ~10% within both groups at both Poor for LDL. No results reported for TC, HDL or TG. 3 and 6 months, (p<0.01). No differences between groups in knowledge, caloric Both groups had increases in knowledge at 3 intake, fat intake. Cholesterol intake was lower in PCAT and 6 months compared to baseline (p< 0.001). group vs. control at 6 mo. (p<0.05) No difference in caloric intake within groups. Total fat intake decreased in both groups; no change in saturated fat intake; both groups decreased cholesterol intake. 4 Table 10. Randomized Controlled Trials of Diet Treatment for Children Without Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Author, year Intervention(s) Control Population Williams, 1995229 Fiber cereal contained 3.2 g soluble fiber per Placebo cereal with 0.5/g of fiber, 1 58 with TC>170 serving. Dose=1 box of cereal per day for 3 box/day x 3 weeks then 2 boxes per day; mg/dL and weeks, then 2 boxes/day. Children ages 2-5 children ages 2-5 consumed 1 box per LDL>110mg/dL. consumed only 1 box/day throughout study. day. Abbreviations AHA=American Heart Association, Apo A=Apolipoprotein A, Apo B=Apolipoprotein B, HDL=High-density lipoprotein, LDL=Low-density lipopr cholesterol, TG=Triglycerides 5 Table 10. Randomized Controlled Trials of Diet Treatment for Children Without Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Quality Author, year Between Group Comparisons Within Group Comparisons Rating Williams, 1995229 TC decreased for fiber vs. Step 1(p<0.05), LDL After 12 weeks: Poor decreased for fiber vs. Step 1 (p<0.01), no difference in Both groups had significant within-group HDL or TG between groups. decreases in TC, but decrease for fiber was TC/HDL and LDL/HDL ratios improved (p<0.001). significantly greater. TC decrease p<0.01 within Step I, p<0.001 within fiber group. No change in HDL within groups. Fiber group had decrease in LDL (p<0.01), No change within step I group. TG decreased for fiber group, with no change within step 1 group. 6 Table 11. Randomized Controlled Trials of Exercise Trials for Children and Adolescents Without Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Duration Between Group Within Group Quality Author, year N of Trial Intervention(s) Control Comparisons Comparisons Rating Obese Children and Adolescents Ferguson, 81 obese 4 months Exercise program offered No Prior to cross-over, no p-For change over 8 Fair 1999225 children each 5 days/week, 40 min per exercise values reported: months (post- mean age period day; children were paid training TC decreased 6% for crossover), group x 9.5 $1/session and given program intervention group and time interaction was prizes for maintaining a 7% for control. significant for HR>150 bpm. After HDL increased 4% for triglycerides only school transportation was intervention vs 0% for (p=0.02) Results provided. control. also given for LDL decreased 6% for combined groups at intervention group vs the end of their 12% for control. intervention periods. TG decreased 17% for intervention vs an increase of 12% for control. Kang, 2002226 80 obese 8 months Physical training (PT) Lifestyle Comparison of pre-post Poor children offered 5 days/week for 8 education changes in mmol/L ages 13- months except when alone. (least-square 16 assigned to lifestyle mean+SEM): education (LSE). No significant change in Physical training = TC or HDL compared to individual exercise control. prescription, 1045 TG decreased for kJ/session (250kcal), LSE+PT group as mean prescribed HR of compared to control 55-60% or 75-80% of (p=0.003). peak 1 Table 11. Randomized Controlled Trials of Exercise Trials for Children and Adolescents Without Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Duration Between Group Within Group Quality Author, year N of Trial Intervention(s) Control Comparisons Comparisons Rating Healthy Children and Adolescents Boreham, 25 7 weeks 7 week stair climbing No change Stair climbers had Pre-post results for Poor 2000230 sedentary program, progressing to in activity significant stair climbers only in but 6 ascents/day on a public improvements mmol/L, mean healthy access staircase Mon-Fri compared to control in +SEM after 7 weeks: female (199 steps with a total HDL (p<0.01) but not TC decreased students vertical displacement of for TC. However, pre- (p<0.05). ages 18- 32.8 meters). intervention difference HDL increased 22 in HDL between stair (p<0.05). Controls climbing and control had no signficant group was significant changes in lipid (p<0.05). levels. Linder, 50 healthy 8 weeks Physical conditioning Usual Post-test physical No significant Fair 1982232 boy scouts program (PA) designed summer working capacity differences in the ages 11- to increase aerobic activities. increased in the PA change in TC, LDL, 17 capacity. group vs control group, HDL, TG, VLDL p<0.05. levels from pre to No difference in TC, post. LDL, HDL, TG or VLDL between groups. 2 Table 11. Randomized Controlled Trials of Exercise Trials for Children and Adolescents Without Identified Familial Hypercholesterolemia or Other Monogenic Dyslipidemia Results Duration Between Group Within Group Quality Author, year N of Trial Intervention(s) Control Comparisons Comparisons Rating Savage, 663 11 weeks Walking/jogging/running Control No significant No significant Fair 1986233 prepube- 3 times/week for 11week subjects differences between changes in HDL/total scent boys (total distance of 1.6 km were groups in the changes cholesterol ratio, TG, per session). Low requested of HDL/total cholesterol or LDL. intensity group had a to maintain ratio, TG, or LDL. training HR of 40% VO2 their Change in HDL-C maximum. High intensity current values was significantly training group had a activity greater for the high training HR of 75% VO2 pattern intensity group (p<0.05) max. throughout but not for the low the study. intensity group as compared to control. Stergioulas, 58 2 months Exercise: 2 month No specific No significant Physical work Poor 1998228 sedentary training program with training differences between capacity of exercise high four, 60 minute sessions program. groups. group increased at school per week. Sessions Both the end of treatment boys ages included warm-up, groups (p<0.001). No 10-14 stretching, and aerobic advised not change seen in normal exercise at 75% of to change control group. body physical work capacity. dietary HDL increased for weight and habits or intervention group at normal physical end of the 2nd lipids activity. treatment month (p<0.005) and at the end of the detraining period (p<0.01). No change seen in control group. Abbreviations HDL=High-density lipoprotein, HR=Heart rate, LDL=Low-density lipoprotein, SEM=Standard error of measure, TC=Total cholesterol, TG=Triglycerides, VLDL=Very low-density Lipoprotein 3 Table 12. Randomized Controlled Trials of Diet and Exercise for Children Without Identified Hypercholesterolemia or Other Monogenic Dyslipidemia Results Author, Duration Between Group Within Group Quality year N of Trial Intervention(s) Control Comparisons Comparisons Rating Becque, 36 over- 20 weeks 1. Diet and behavior No Post-treatment HDL level Control group had no Fair 1988221 weight change: met with dietician change for exercise, diet and pre-post differences children and behavior therapist in behavior change group in HDL, TC or TG ages. once/week. ADA exchange activity was significantly better (n=14). 3 small program set to elicit the loss or diet from the control and from groups. 1-2 lbs/week. the diet plus behavior Diet and behavior Behavioral treatment change post-treatment change group had no included record keeping groups (p<0.05) pre-post differences stimulus control, changing No other between group in TC, HDL or TG topography of eating and differences. (n=11). reinforcing altered behavior. Exercise, diet and 2. Exercise plus diet and behavior change behavior change: same as group increased HDL above, with exercise (P<0.05) but did not program 50 minutes 3 change TC or TG days/week. (n=11). Epstein, 56 obese 6 months Diets were set between Results for the two Poor 1989217 (>20% of 3800-5000/kJ/day and intervention groups (diet ideal monitored by a nutritionist to or diet+exercise weight) maintain nutrient adequacy. instruction) were children Information on diet, combined for total ages 8- exercise, stimulus control, treatment group n=35 12 reinforcement, modeling and and control n=16. contingency contracting was TC decreased for presented to parents and intervention vs. control their children in eight weekly (p=0.03). sessions followed by four HDL increased for monthly sessions. intervention vs. control, (p=0.007). TG decreased for intervention vs. control (p=0.01). 1 Table 12. Randomized Controlled Trials of Diet and Exercise for Children Without Identified Hypercholesterolemia or Other Monogenic Dyslipidemia Results Author, Duration Between Group Within Group Quality year N of Trial Intervention(s) Control Comparisons Comparisons Rating Walter, 1115 1 school "Know Your Body" Usual Adjusting for age, At year 1: Fair 1985222 4th year (2 curriculum yearly from 4th- cur- gender, race and TC decreased 0.4% graders hours/ 8th grade; taught in usual riculum baseline risk factor level, in the intervention week) classroom by regular TC difference between group vs. a 0.5% teacher 2 hours/week for groups was significant increase in control entire school year. (p=0.03). HDL and group. Teachers trained by HDL/TC differences were HDL decreased 7% research staff in 3 half-day not significant (p=0.18 in intervention group teacher workshops on and p=0.06 respectively). vs 9% in control curriculum implementation. group. Adherence to teaching TC/HDL ratio protocols monitored by decreased 9% for classroom visits from intervention vs 6% in research staff and control group. attendance at training workshops. Abbreviations ADA=American Dietetic Association, HDL=High-density lipoprotein, TC=Total cholesterol, TG=Triglycerides 2 Table 13. Adverse Events Reported in Studies of Statins Study Age Author, year, title Design N Duration (years) Adverse Effects of Treatment Laboratory Values Atorvastatin McCrindle, 200376 RCT 187 26 weeks 10-17, No significant adverse effects; No Increased AST in 1%; Increased ALT in mean effect on sexual development. 1%. None withdrew or stopped med as 14.1 a result of increased transaminases. Lovastatin Clauss, 200538 RCT 54 girls 24 weeks 10-17 None observed; no detectable NR effect on menstrual cycle length. Lambert, 199672 RCT 69 boys 8 weeks <18, None observed. Asymptomatic elevations in CK in 3 mean subjects. 12.7 1 Table 13. Adverse Events Reported in Studies of Statins Study Age Author, year, title Design N Duration (years) Adverse Effects of Treatment Laboratory Values Stein, 199978 RCT 132 48 weeks mean 13 No effect on growth; No difference DHEAS increased by 18%. Tocopheral between drug and placebo in decreased. ALT increased in both drug testicular volume, Tanner staging, and placebo; no differences between testosterone, or LH level. groups at week 48. No consistent changes in AST or CK. Transient CK elevations in response to exercise. Mean CD3 lymphocyte count decreased from baseline of 2.72 + 0.91 nX109 to 6-months of 2.34 + 0.81 nX109 (p=0.003). Mean CD4 lymphocyte count decreased from baseline of 1.47 + 0.57 nX109 to 6- months of 1.26 + 0.50 nX109 (p=0.010). Mean CD8 lymphocyte count decreased from baseline of 0.89 + 0.34 to 6-months of 0.75 + 0.26 nX109. Pravastatin Wiegman, 200437 RCT 214 2 years 8-18, No impact on growth, endocrine No effects on muscle or liver enzyme mean 13 function parameters, Tanner levels. staging scores, onset of menses or testicular volume. Hedman, 2003240 Open-label 20 8 weeks 4-15 Abdominal pain (1), loose stools No effects on serum ALT, CK, or (1), headache (4), sleep creatinine. disturbance (2), muscle tenderness or pain at rest (1), muscle tenderness or pain associated with physical training (1). 2 Table 13. Adverse Events Reported in Studies of Statins Study Age Author, year, title Design N Duration (years) Adverse Effects of Treatment Laboratory Values Knipscheer, 199671 RCT 72 12 weeks mean 12 Rash, nose bleeding, headache, CK abnormal in 8 of placebo, 6 of 5 nausea/vomiting, abdominal pain. mg/day, 11 of 10 mg/day and 8 of 20 mg/day groups. Cortisol abnormal in 2 of placebo, 2 of 5 milligrams per day, 5 of 10 milligrams per day, 3 of 20 milligrams per day groups. Simvastatin Couture, 199866 RCT 63 6 weeks 8-17, None observed. NR mean 12.6 De Jongh, 200268 RCT 69 28 weeks 9-18, None observed. No significant differences between mean simvastatin and placebo FH groups on 14.6 ALT, AST, and CK. De Jongh, 200269 RCT 173 48 weeks 10-17 Abdominal pain (3), chest pain (1), Increased ALT (3), AST (3), and CK flatulence (1), myalgia (2), (1). headache (4), sleep disorder (1), weight gain (1), pruritus (1). Dirisamer, 2003241 Open-label 20 18 10-17, Transient headache (2). Myalgia Slightly higher values of CK (2); months mean 13 (1) for 2 weeks. Transient Transiently elevated ALAT (GPT) and gastrointestinal complaints (2). GCT (1). 3 Table 13. Adverse Events Reported in Studies of Statins Study Age Author, year, title Design N Duration (years) Adverse Effects of Treatment Laboratory Values Ducobu, 1992243 Open-label 32 24-36 <17 Children for whom height (n=12) Slight transaminase increase (n=1) and months and weight (n=16) were available slight CK increase (n=2) were transient. remained in the growth No significant changes detected percentages they had been in at overall. baseline. Stefanutti, 1999244 Non-RCT 16 12 7-12, No adverse effects were observed NR months mean with diet alone or with diet plus 8.75 simvastatin. Patients showed good compliance with the treatment. Various or unspecified statins Sinzinger, 2004242 Descriptive 22 8 years 15-27, Muscle pain reported in 84% of Elevated CK in 3 subjects. No profes- mean periods of statin therapy. Mean increase in liver enzymes. sional 24.1 time of onset of muscle pain was athletes 8.3 days. De Jongh, 200370 Cross- 69 NR 10-18, No differences between the Not assessed. sectional mean children with FH and their healthy 15.3 peers on health-related quality of life and anxiety. Abbreviations ALAT=Alanine aminotransferase, ALT=Alanine transaminase, AST=Aspartate aminotransferase, CK=Creatine kinase, DHEAS=Dehydroepiandrosterones, GCT=Glucose challenge test, GPT=Glutamic-pyruvic transaminase, FH=Familial hyperlipidemia, LH=Luteinizing hormone, NR=Not reported, RCT=Randomized controlled trial 4 Table 14. Adverse Events Reported in Studies of Bile-acid Binding Resins Age Author, year, title N Duration (years) Adverse Effects of Treatment Laboratory Values Cholestyramine Curtis, 1991245 1 2 years 7 Loss of dental enamel noted. This was Serum calcium, phosphorus, folate, B12 presumed due to low pH (2.4) of were normal cholestyramine mixed with Kool-Aid (and swished by child in his mouth for 15 min before swallowing) Farah, 1977246 20 16 days 4-23, Febrile gastroenteritis in 1 patient after Serum folate decreased significantly in mean 15 7 days treatment resulting in females. 7/11 females had folate discontinuation of therapy. measured after treatment and results Farah, 1977247 only reported in aggregate by gender. SGOT increased in 2 patients (32 and 41 IU/dL), and persisted 6 months. Transient LDH increases (>60 IU/dL) in 2. No fat-soluble vitamin malabsorption. Glueck, 1973263 20 on diet 6 months with 7-21 10 of 12 routinely took the suggested No consistent changes in any of the +BABR additional 12- dose of cholestyramine. laboratory safety tests (CBC, liver 18 months Weight gain and growth progressed function tests, BUN, calcium, follow-up for during the 12 and 18 month follow-up phosphorous, total protein, electrolytes growth period along normal growth percentile and carbon dioxide). No evidence for hypercholoremic acidosis at any time. Glueck, 1974262 30 on diet average 5-21 NR Plasma vitamins A and E remained + BABR follow-up of 6 within the normal range and paralleled months at higher levels the 25th to 75th percentile distirbution for vitamin A and E in normal children. 1 Table 14. Adverse Events Reported in Studies of Bile-acid Binding Resins Age Author, year, title N Duration (years) Adverse Effects of Treatment Laboratory Values Glueck, 1977260 16 (same 16 children 9-17 11 of 16 had good adherence. 5 CBC, liver function tests, vitamin A and children followed for children dropped out after 2 years E, calcium, phsophorus, blood urea as Glueck 18 months; because of poor palatability of the nitrogen, fasting blood sugar levels did 1973 12 for 24 drug. not vary signficantly over time when above) months; 7 for 1 child had persistent constipation. compared to baseline levels on diet 30 -36 None had nausea. 5 complained that only. months resin was gritty, had poor palatability. 1 complained of chronic fatigue Glueck, 1986261 33 4.3 years mean Height and weight percentiles at NR 10.3 completion were unchanged from baseline. No abnormal growth patterns noted. Sexual maturation was normal. None had unusual infections or extra school sick days. None of the girls had ammenorrhea; though 1 competitive cross-country runner had persistently irregular periods. No depression, suicide attempts or traumatic deaths. 4 children exhibited transient refusal to follow diets and/or take bile acid- binding resins. Koletzko, 1992251 35 on Diet: mean 2-17, No serious side effects were noted NR diet; 14 17.5 mos. mean 7.9 with both forms of treatment (diet/ diet on diet + Diet+BABR: + BABR). Normal growth. Mild BABR mean 27.9 gastrointestinal symptoms occurred in months "very few" patients. Liacouras, 1993252 87 Up to 62 mean 12 nausea, 2 abdominal bloating, 1 No elevated prothrombin times. months 10.6 severe constipation. 73% complained of poor palatability 2 Table 14. Adverse Events Reported in Studies of Bile-acid Binding Resins Age Author, year, title N Duration (years) Adverse Effects of Treatment Laboratory Values McCrindle, 199774 40 28 weeks 10-18 Minor gastrointestinal complaints were NR frequent but did not result in any drop- out Tonstad, 199681 96 1 year 6-11 Growth and bone age were normal. Folate deficiency occurred in most mean 8 One case of intestinal obstruction subjects with cholestyramine, and caused by adhesions in 1 patient. vitamin D levels decreased significantly Unpalatability, headaches, and for those not taking a multi-vitamin. No vomiting were reasons for withdrawals. other significant differences in nutrient levels. Hemoglobin and liver enzymes were unchanged. Tonstad, 1998255 96 1 year 6-11 22/36 treatment and 26/36 placebo During cholestyramine treatment, mean 8 participants completed the trials plasma total homocysteine increased in thought secondary to unpalatability of subjects with the C677T mutation in one drug. or both alleles but not in subjects with the CC genotype. West, 1973256 19 Up to 20 Some had impaired fat absorption Serum folate decreased in all patients. months without diarrhea. Growth was normal. Hemoglobin, prothrombin time, serum None developed anemia. calcium, alkaline phosphates, and serum vitamin A concentrations were unchanged. West, 1975257 18 1 to 2.5 years 1-14 No child developed diarrhea, and Folate deficiency and decreased red growth was normal. cell folate. Prothrombin time remained normal in all patients. There was a significant decrease in mean serum levels of vitamins A and E and inorganic phosphorus. Levels of serum iron, vitamin B12, plasma calcium, and protein did not change significantly. West, 1975258 45 2-8 years 1-16 Adherence was poor due to Folate deficiency, steatorrhoea, and unpalatability of cholestyramine. reduction in serum levels of vitamins A and E and of inorganic phosphorus although not to abnormally low values. 3 Table 14. Adverse Events Reported in Studies of Bile-acid Binding Resins Age Author, year, title N Duration (years) Adverse Effects of Treatment Laboratory Values West, 1980259 35 1-8 years 1-17, Almost all children expressed some NR mean 8.4 dislike of cholestyramine, and a few complained of transient gastric fullness. No patient had persistent constipation, loss of appetite or diarrhea. Nausea, dizziness and malaise in a female aged 18. One boy died of intercurrent infection 10 months after starting meds, not stated whether related to treatment. 4 Table 14. Adverse Events Reported in Studies of Bile-acid Binding Resins Age Author, year, title N Duration (years) Adverse Effects of Treatment Laboratory Values Colestipol Groot, 1983248 33 16 weeks NR 5 withdrew due to unpalatability. NR Hansen, 1992249 30 8.5 years 1-17 One child's height/age decreased NR (diet); 5.5 below -2 SD. Growth was normal in years (diet other children followed by diet + BABR) Harvengt, 1976250 3 Up to 36 6, 11, 18 No hepatic dysfunction, renal function, Low iron without anemia in one. Serum months or hematologic side effects observed. uric acid level increased during Mild gastrointestinal complaints treatment but did not reach abnormal (flatulence, constipation) during first 3 values. Serum vitamin B12 and folic months, but disappeared despite acid content were not impaired after 2 continued treatment. No steatorrhea. and 3 years on colestipol. Prothrombin time values were not abnormally prolonged. McCrindle, 200275 40 36 weeks 9-18, 18% reported constipation, 21% had NR median stomachache, 11% had headache, 6% 14 had muscle aches. Compliance was 57-66% overall and was similar among treatment groups and dose groups. 5 Table 14. Adverse Events Reported in Studies of Bile-acid Binding Resins Age Author, year, title N Duration (years) Adverse Effects of Treatment Laboratory Values Schwarz, 1980253 23 Up to 24 5-17, 6 subjects complained of poor Serum vitamins A and E decreased months mean 12 palatability. 1 patient developed significantly after 18-24 months of Reynauld's phenomenon during colestipol. No changes in 25- therapy but continued treatment hydroxycholecalciferol, folic acid or without recurrence of symptoms. prothrombin time. Tonstad, 199680 66 52 weeks mean 8 had gastro-intestinal side effects, Low-dose colestipol (10 g/day) reduced 13.2 including constipation, dyspepsia, concentrations of serum folate after 8 flatulence, nausea, decreased weeks. Serum vitamin E and appetite, abdominal pain. Growth was carotenoids decreased proportionally normal. with decreases in cholesterol. Vitamin D did not change significantly, but decreased more in subjects who were more compliant after 1 year. Tonstad, 1996254 27 6 months for 10-16 Growth was normal. Two had difficulty NR colestipol; swallowing the tablets; one had mean 6 years difficulty with flatulence; one had for diet abdominal discomfort. Abbrevations BABR=Bile acid binding resin, BUN=Blood urea nitrogen, CBC=Complete blood count, LDH=Lactate dehydrogenase, NR=Not reported, SGOT=Serum glutamic-oxalocetic transaminase 6 Table 15. Adverse Events Reported in Studies of Other Drugs and Combinations Study Age Adverse Effects of Compliance/ Author, year, title Drug Design N Duration (years) Treatment Laboratory Values Tolerance Baker, 198254 Probucol Open- 7 15-21 6-21 Nausea in one patient; No significant changes in Well tolerated label trial months Growth and development endocrine, hepatic, or were normal. hematological investigations. Levels of total T1 and T4, and of free T1 and T4 did not change. Becker, 1992264 Sitosterol Open- 7 3 months mean Two had decreased Sitosterol: slight, significant Children and and label trial sitosterol; 3 8.4 appetite for the first 2 decrease in hemoglobin parents reported a bezafibrate, months weeks on sitosterol. (-5%) and ALP (-19%). high degree of in sequence bezafibrate; Ultrasounds of Bezafibrate: ALP remained acceptance and and in 24 months gallbladder found no lower; iron increased by 26%. compliance during combination sitosterol + abnormalities. Combination: transferrin all treatments. bezafibrate increased 20% and reached abnormal levels in 2; all other lab values normal. Brun, 1980265 Dextrothy- Descrip- 6 12 weeks 7-12 NR With D-T4 treatment, the NR roxine tive secretion of both TSH and T3 (D-T4) in response to TRH was abolished. An increase in the basal level of T3 was observed after treatment with D-T4. The high circulating levels of D-T4 and possibly of D-T3 after chronic administration of D-T4 may be responsible for the saturation of pituitary nuclear T3 receptors, resulting in the suppression of the TRH- induced TSH response. 1 Table 15. Adverse Events Reported in Studies of Other Drugs and Combinations Study Age Adverse Effects of Compliance/ Author, year, title Drug Design N Duration (years) Treatment Laboratory Values Tolerance Colletti, 1993266 Niacin Retro- 21 1-19 4-14 18 of 21 patients Reversible serum 18 of 21 patients spective months, reported some adverse aminotransferase elevations reported some average 8.1 effect: flushing 71%, (dose related) in 6 patients: 4 adverse effect. itching 19%, abdominal with crystalline and 2 with Drug was pain 14%, nausea 14%, sustained release form of discontinued in 1 headache 14%, niacin. subject due to constipation 5%. One poor compliance. subject developed febrile illness with serum aminotransferase >400 IU/L two months after starting niacin. Niacin at 1000 mg/day-sustained release was considered a possible cause of hepatitis. Malloy, 197873 P-amnosali- RCT 20 6 months 5-21, No indicators of drug Normal SGOT, SGPT, ALP, Compliance was cylic acid mean toxicity reported, other bilirubin, and glucose levels >50% for 11 12 than mild gastric irritation in fasting serum; normal TSH subjects, <50% for that remitted with oral and thyroxine. 9 subjects. antacid treatment. Triiodothyronine levels were slightly below normal for 2 in active drug and 1 on placebo. 2 Table 15. Adverse Events Reported in Studies of Other Drugs and Combinations Study Age Adverse Effects of Compliance/ Author, year, title Drug Design N Duration (years) Treatment Laboratory Values Tolerance McDuffie, 2002267 Orlistat Observa- 20 3 months mean Effects were generally A small but significant drop in Subjects who tional 14.6 mild, limited to 25-hydroxy vitamin D levels completed gastrointestinal effects was seen at 1 month. 3 treatment (85%) related to increased fat subjects required additional reported taking excretion, and resolved vitamin D supplementation 80% of prescribed within the first 6 weeks of despite the prescription of a medication. treatment. daily multivitamin containing Only one subject vitamin D. TSH, free (5%) cited thyroxine, glycosylated intolerance of hemoglobin, calcium, adverse effects as phosphorous, magnesium, reason for zinc, measures of iron stores withdrawal. did not change significantly. Stein, 1989268 Diet + drug Cross- 30 1-9 years 1-20, No adverse clinical or Resin + niacin together NR or combined sectional mean biochemical side effects produced elevated AST and drugs: 5.5 were noted on the HMG ALT and clinical symptoms of BABR; CoA reductase inhibitor. hepatotoxicity in one subject. BABR + Niacin in this subject was niacin; associated with a significant lovastatin or rise in liver enzymes, simvastatin suppression of albumin, and clinical symptoms of hepatotoxicity. 3 Table 15. Adverse Events Reported in Studies of Other Drugs and Combinations Study Age Adverse Effects of Compliance/ Author, year, title Drug Design N Duration (years) Treatment Laboratory Values Tolerance Steinmetz, 1981269 Fenofibrate Descrip- 17 18 months 4-19 NR 4 of 17 subjects had NR tive increased ALT (at least 100%) and AST (at least 80%). Uric acid decreased significantly (mean change - 20%). Bilirubin decreased significantly (mean change - 19%). Inorganic phosphates decreased slightly but significantly. Albumin was unchanged. ALP decreased by mean 15%. GGT decreased slightly (by 1.5 U/l). ALT increased by 9 U/l. AST increased by 11 U/l. Wheeler, 198582 Bezafibrate RCT 14 3 months 4-15, Growth was satisfactory One child had high alk phos All subjects mean throughout trial. No other after 3 months of bezafibrate. declared 10.9 reports of adverse effects Another child had slight preference for this transient rise in ALT during drug over first 2 months on bezafibrate cholestyramine but values were normal by used previously. end of 3rd month. Abbreviations ALP=Alkaline phosphate, ALT=Alanine transaminase, AST=Aspartate aminotransferase, BABR=Bile acid binding resin, NR=Not reported, SGOT=Serum glutamic-oxalocetic transaminase, SGPT=Serum glutamic-pyruvic transaminase, TRH=Thyroid releasing hormone, TSH=Thyroid stimulating hormone 4 Table 16. Adverse Events Reported in Studies of Low Fat Diet Study Age Adverse Effects of Author, year, title Design N Duration (years) Growth Treatment Laboratory Values Cetta, 1994279 Retro- 63 2 years 2-16, Normal height. Weight NR NR spective mean loss in 2 females: 1 with 7.8 anorexia nervosa, 1 with a more restrictive vegetarian diet. Copperman, 1995270 Cross- 54 NA mean NR NR Normal intake of energy, sectional 10.8 minerals, or vitamins D and E. DISC Collaborative RCT 663 3 years mean Normal NR Intake was low for vitamin E Research Group, 1995219 9.4 and zinc. Feoli-Fonseca, 1998282 Retro- 16 23 years infants Normal 2 children had severe Abnormal values were spective reviewed; (<1 abdominal pain, and observed for serum iron, mean follow- year) another had acute alkaline phosphatase, and total up 6.76 years with pancreatitis after using calcium. All erythrocyte folate familial an oral contraceptive and albumin levels were within chylo- agent. normal range. micro- nemia 1 Table 16. Adverse Events Reported in Studies of Low Fat Diet Study Age Adverse Effects of Author, year, title Design N Duration (years) Growth Treatment Laboratory Values Jacobson, 1998280 Observa- 138 3 years 2-15 Normal NR NR tional Kaistha, 2001285 Cross- 80 N/A 9.9 NR NR More children with sectional hyperlipidemia than controls (p<0.05) consumed below 75% of the RDA/DRI for vitamin E and calories. Kuehl, 1993227 RCT 295 33 weeks mean 7 Normal NR Iron intake maintained at over 87% and calcium intake at over 81% of the RDA throughout the study. Lavigne, 1999281 RCT 663 Mean follow- 8-10 NR No adverse effects of NR up 36.2 treatment in academic months function, psychological symptoms, or family function. 2 Table 16. Adverse Events Reported in Studies of Low Fat Diet Study Age Adverse Effects of Author, year, title Design N Duration (years) Growth Treatment Laboratory Values Lifshitz, 1989278 Observa- 40 Mean time mean 8 (20%) of 40 patients NR The 3 patients with nutritional tional since 7.7 had growth failure. dwarfing consumed <60% of diagnosis: Three had nutritional required energy, <50% RDA of 20.1 months dwarfing with no vitamins B6, D, folacin, zinc, (with growth progression of puberty. iron, calcium, and magnesium, failure); 3.9 The other 5 had a drop in and <66% of RDA for vitamins months body weight without A and B12, niacin, and (without linear growth alterations. phosphorus. Only vitamins E growth and C were adequately failure) consumed. McKenzie, 1996274 Prospec- 300 3 months 4-10 NR NR 74% failed to consume two tive thirds of the RDA for vitamin D. cohort Moreno, 1998275 Descrip- 42 6 months 7-15 NR NR Lymphocyte T subset counts tive (CD3, CD4, and CD8) showed significant decreases after 6 months but were within normal ranges. 3 Table 16. Adverse Events Reported in Studies of Low Fat Diet Study Age Adverse Effects of Author, year, title Design N Duration (years) Growth Treatment Laboratory Values Obarzanek, 2001218 RCT 663 4 years mean Normal height, BMI, and NR No abnormalities in serum 9.5 sexual maturation ferritin, red blood cell folate, serum retinol and zinc. Rose, 1976277 Descrip- 16 16 followed 1 month Normal None observed NR tive for 1 year; 9 to 20 for >1 year years Sanchez-Bayle, 1994276 Observa- 451 6-24 months 2-18 Normal NR NR tional Sanchez-Bayle, 2003283 Prospec- 144 Mean 7.42 2-13, Normal height. Increase NR NR tive years mean in weight 5.5 4 Table 16. Adverse Events Reported in Studies of Low Fat Diet Study Age Adverse Effects of Author, year, title Design N Duration (years) Growth Treatment Laboratory Values Segall, 1970272 Observa- 5 10 days 6-15 1 child lost 1.4 kg, NR In all subjects, serum tional another child gained 1.2 triglyceride increased after 3-6 kg days on the high-carb diet. TG levels fell on resumption of a diet containing 45% carbohydrates. Tershakovec, 1998284 RCT 261 12 months 3.9-9.9 Normal None noted. NR Tonstad, 1996271 Cross- 185 N/A 7-16 Normal Normal behavioral and NR sectional emotional scores. Witschi, 1978273 Observa- 91 3 week diet Adoles- Normal NR NR tional period cents Abbreviations BMI=Body mass index, DRI=Dietary reference intakes, NA=Not applicable, NR=Not reported, RCT=Randomized controlled trial, RDA=Recommended daily allowance, TG=Triglycerides 5 Table 17. Adverse Events Reported in Studies of Dietary Supplements Dietary Study Age Adverse Effects of Compliance/ Author, year, title Supplement Design N Duration (years) Treatment Laboratory Values Tolerance Amundsen, 200241 Plant sterol RCT 38 8 weeks mean None observed. ALT increased 16.8% (p = Children (some data from ester spread 10.5 0.04). After adjustment for lipid consumed 18 Amundsen 2004) 18.2 g/day of changes, lycopene was 8.1% g/day of spread. spread lower (p=0.015) in treatment 1 subject corresponding than in control group. lipid- withdrew to 1.6 g/day of adjusted serum retinol levels because the sterol esters. were 15.6% (p<0.001) higher in amount of spread treatment than control groups. was too large. Lipid-adjusted lathosterol levels increased by 96% ; sitosterol increased by 48% Amundsen, 2004287 Plant sterol Open- 37 26 weeks mean NR Lipid-adjusted lathosterol levels Children ester spread label 9.6 were stable; sitosterol levels consumed (20 g of trial, were 77% higher compared to slightly less spread per following control period of RCT. spread (14 g/day day, RCT Lipid-adjusted retinol levels vs 18 g/day corresponding above were 11% higher at the end of during trial). to 1.76 g/day the open label period as of plant compared to control. No sterol), some changes in lycopene or lutein. children took Lipid-adjusted serum a-carotene fish oil increased in the open label supplements period following a decrease in or vitamins the trial period. 1 Table 17. Adverse Events Reported in Studies of Dietary Supplements Dietary Study Age Adverse Effects of Compliance/ Author, year, title Supplement Design N Duration (years) Treatment Laboratory Values Tolerance Becker, 1993288 Diet plus Open- 9 10 months: mean Growth was normal. Diet phase: ALT decreased. All 9 subjects sitosterol label trial 3 sitosterol 11.9 No abnormalities in Sitosterol: alk phos and completed the pastils, then 7 sitostanol liver and gallbladder. carotene decreased. Sitostanol: study; good diet plus No changes in stool ALP and carotene returned to compliance with sitostanol pattern. initial values; transferrin all treatments. increased slightly and serum bile acid levels decreased, but were within normal range. No changes in hemoglobin, leukocyte count, platelet count, calcium, phosphate, lipase, iron, creatinine, and creatinine kinase. Clarke, 1990289 Fish oil 5 Observa- 11 6 months 11-21 8 of 11 subjects had No evidence of liver 2 (18.2%) of 11 g/day tional episodes of epistaxis dysfunction. All prothrombin and subjects withdrew that caused 2 to partial thromboplastin times and due to epistaxis. withdraw. One platelet counts were normal. subject had asymptomatic occult blood in the stool on one occasion; the bleeding time was normal. 2 Table 17. Adverse Events Reported in Studies of Dietary Supplements Dietary Study Age Adverse Effects of Compliance/ Author, year, title Supplement Design N Duration (years) Treatment Laboratory Values Tolerance Dennison, 1993223 Psyillium fiber RCT 25 4-5 weeks mean No changes in sub None of vitamin or mineral Compliance was 6 g/day 11.1 scapular or triceps levels tested were decreased. 82% for both skin-fold thicknesses. cereals. Height and weight growth were normal. Glassman, 1990290 Psyillium fiber Descrip- 36 8.1 months 9-17, No abdominal No changes in serum zinc, All patients 5 g/day tive mean distention or copper, or hemoglobin tolerated both 9.7 cramping, concentrations; hematocrit; regimens well. constipation, mean corpuscular volume; or diarrhea, or excessive prothrombin and partial flatus production. thromboplastin times. Gulesserian, 2002291 Rapeseed oil Descrip- 17 5 months 4-19, NR NR The diet was well substitution for tive median accepted; no all fats and 12.7 patient withdrew food due to dislike of preparation the oil. 3 Table 17. Adverse Events Reported in Studies of Dietary Supplements Dietary Study Age Adverse Effects of Compliance/ Author, year, title Supplement Design N Duration (years) Treatment Laboratory Values Tolerance Gylling, 199539 Rapeseed oil- RCT 14 6 weeks 2-15, NR NR Reported good rich margarine HeFH, mean compliance; well substitution of 1 9.1 tolerated. fats (24 g/day) HoFH Children noted with sitostanol no difference in (3 g/day) taste between the two margarines. Laurin, 1991294 Soy-protein RCT 10 4 weeks mean Growth and NR Reported good beverage 7.9 development were control over normal. mean dietary intake; similar percentages of protein energy were consumed as a dairy source in the soy-protein and cow-milk diet groups. 4 Table 17. Adverse Events Reported in Studies of Dietary Supplements Dietary Study Age Adverse Effects of Compliance/ Author, year, title Supplement Design N Duration (years) Treatment Laboratory Values Tolerance McCrindle, 1998235 Garlic extract RCT 30 8 weeks 8-18, The most common Serum albumin level (+2.0 g/L, 1 patient was mean effects were p=0.002) and hemoglobin (+5.2 unable to swallow 14 headache and upset g/L, p=0.02) with garlic. the capsules and stomach, and were withdrew. similar between garlic Compliance was and placebo groups. similar between 1 patient on garlic garlic and had unpleasant body placebo groups. odor, but had been 86% vs 93% working (p=0.34). concomitantly on a garlic farm during the study. Mietus-Snyder, Vitamins E Non- 11 6 weeks 6-21, No adverse effects of NR All subjects 1998292 (400 IU) and random- mean antioxidant vitamin tolerated the C (500 mg) ized 12.5 therapy reported. study well. twice daily controlle d trial Sanchez-Bayle, Fiber tablets Observa- 53 3 months 4-18, Two children reported NR Generally well 2001293 (50% wheat tional mean abdominal discomfort tolerated; 2 bran, 50% 7.3 and soft stools. withdrew due to pectin) 100- abdominal 150 discomfort mg/kg/day 5 Table 17. Adverse Events Reported in Studies of Dietary Supplements Dietary Study Age Adverse Effects of Compliance/ Author, year, title Supplement Design N Duration (years) Treatment Laboratory Values Tolerance Schlierf, 1978295 Sitosterol RCT 15 3 months 8-20 Not reported. NR Good compliance. Zavoral, 1983296 Locust bean Non- 11 8 weeks 10-18 Increased rectal gas No change observed in the Children ate the gum (LBG) RCT for 1 to 2 weeks while white blood cell counts, serum prescribed cross- on the LBG- calcium, or SGOT amount of LBG over trial containing products. determinations. food products. Abbreviations ALP=Alkaline phosphatse, ALT=Alanine transaminase, HeFH=Familial hyperlipidemia, HoFH=Familial homozygous hyperlidpidemia, NR=Not reported, RCT=Randomized controlled trial, SGOT=Serum glutamic-oxalocetic transaminase 6 Table 18. Summary of Systematic Evidence Review Level and Type Quality of Arrow Key question of Evidence Evidence Conclusions 1 Is screening for dyslipidemia None in children effective in delaying the onset and reducing the incidence of CHD-related events? 2 What is the accuracy of See below. screening for dyslipidemia in identifying children at increased risk of CHD-related events and other outcomes? 2a What are abnormal lipid II-2 Fair to Poor Normal values for lipids in children are currently defined values in children? according to population levels (percentiles). NCEP recommendations are based on LRC data, which defines the 95th percentile for TC as 200 mg/dL and for LDL as 130 mg/dL. There are more recent studies suggesting that age, gender, racial differences and temporal trends shift these cut points. The NCEP has defined levels of LDL for which drug treatment (LDL>190mg/dL or LDL>160mg/dL with family history of early CHD), further evaluation, diet therapy and testing (LDL>130mg/dL) and diet therapy with increased surveillance (LDL110-129mg/dL) are recommended. 2b What are the appropriate II-2, II-3. Poor The most appropriate test is one that accurately predicts future tests? How well do screening risk and benefit from treatment. In the general population of tests (non-fasting total children there have not been adequate studies to determine cholesterol, fasting total these characteristics. Data from few studies suggest that TC cholesterol, fasting lipid above the 95th percentile predicts LDL above the 95th panel) identify individuals with percentile with 44-69% sensitivity. TC minus HDL might be a dyslipidemia? more sentivitive test, but has not been extensively evaluated. A single TC measurement is inadequate to classify children and adolescents into NCEP risk categories with 95% confidence. 1 Table 18. Summary of Systematic Evidence Review Level and Type Quality of Arrow Key question of Evidence Evidence Conclusions 2c How well do lipid levels track II-2. Longitudinal Good Serial correlations measured in individual children over time from childhood to adulthood? cohort studies. are higher for TC (r=0.38-0.78) and LDL (r=0.4-0.7) than for HDL and TG. Approximately 40-55% of children with elevated lipids (by percentile) will continue to have elevated lipids on follow-up as older children. In two studies that followed children/adolescents to young adulthood (between ages 20- 30), 40-45% continued to have elevated lipids. 2d What is the accuracy of II-2. Studies of Good Multiple good quality studies evaluating the use of family history family history in determining diagnostic test as a diagnostic test for dyslipidemia in children using varied risk? accuracy. and large populations demonstrate that family history is an imperfect screening tool for detecting dyslipidemia among children. 2f What are other important risk II-2. Cross- Good for family Evidence from epidemiologic cross-sectional and cohort factors? sectional and history; Good for studies establish a statistical association between family history longitudinal obesity; Poor for and overweight and elevations in lipids. There is inadequate cohort studies. all other risk evidence to show the magnitude of the effect of overweight on factors. lipids, or the impact that incorporating weight measures into a screening tool could have. Multiple other risk factors (diet, physical inactivity, aerobic capacity/fitness, puberty level and smoking) have not been evaluated adequately to assess their contribution to dyslipidemia in children or their usefulness as screening tools. 2e What are effective screening I, III. One RCT Poor Currently recommended screening strategies have limited strategies for children and multiple diagnostic accuracy, low adherence to guidelines by providers, (including frequency of descriptive and limited compliance by parents and children. No trials testing, optimal age for studies of compare strategies of screening in children. No studies testing)? screening address the frequency and optimal age for testing. programs. 2 Table 18. Summary of Systematic Evidence Review Level and Type Quality of Arrow Key question of Evidence Evidence Conclusions 3 What are the adverse effects I, II-3, III. Fair Studies demonstrate lack of parental compliance with of screening including false screening and follow-up recommendations. Reasons for non- positives, false negatives, compliance include concern about test accuracy, lack of proof labeling, etc? that intervention makes a difference in children, concern about upseting the child, refusal by the child, inconvenience, or decision to institute a diet themselves and have child rechecked subsequently. 4 In children and adolescents, None what is the effectiveness of drug, diet, exercise, and combination therapy in reducing the incidence of adult dyslipidemia, and delaying the onset and reducing the incidence of CHD-related events and other outcomes (including optimal age for initiation of treatment)? 5-8 What is the effectiveness of I. RCTs in Good quality Statins are effective for reducing TC and LDL in children with drug, diet, exercise, and familial studies with fair familial hypercholesterolemia. It is not clear how this efficacy combination therapy for hypercholesterol external validity translates to children with milder and/or non-familial forms of treating dyslipidemia in emia patients for drug therapy. dyslpidemia. Diet supplements (psyllium, oat, sterol children/adolescents (drugs and diet) Fair to poor for margarine) and counselling were marginally effective in both (including the incremental and general diet and exercise FH/FCH children and adolescents and those without identified benefit of treating populations of treatments. monogenic dyslipidemia. Exercise treatments showed minimal dyslipidemia in childhood)? children (diet to no improvements in children without monogenic and exercise) dyslipidemia. 3 Evidence Table 1. Tracking of Serum Lipid Levels Study, year arranged by Duration of Fasting serum cohort N Demographics follow-up Age at start Age at follow-up levels? Bogalusa Heart Study; Bogaulsa, LA Bao, 199692 1169 64% white, 36% black, 15 years but dates 5-14 years (calc) Yes 50% M; 50% F given are 15-17 years post start Freedman, 198517 1052 49% M; 51% F 8 years 2.5-14 years Yes 55% white, 45% black Berenson, 197993 1052 212 for TC; 154 for LDL, 2 years 5 years 7 years Yes HDL and TG 1,052 281 for TC; 247 for TG; 2 years 8 years 10 years Yes 245 HDL and LDL 1,052 323 for TC; 270 for LDL, 2 years 11 years 13 years Yes HDL and TG 1,052 241 for TC; 200 for LDL, 2 years 14 years 16 years Yes HDL and TG Child and Adolescent Trial for Cardiovascular Health Cohort Study; Multi-centered in U.S. Kelder, 20023 3,659 at grade 8 (3rd 51% M, 49% F; 71% 6 years 3rd grade (mean 8th grade (mean No follow-up point); HDL White, 13% AA, 13% age 8.8 years) age 14.1 years) done on a random Hispanic, 3% other 30% of population 4,019 same 2 years 3rd grade (mean 5th grade (mean No age 8.8 years) age 11.1 years) 1 Evidence Table 1. Tracking of Serum Lipid Levels Correlation coefficients Study, year arranged by Other Adjusted? If so, cohort TC LDL HDL TG Outcome for what? Comments Bogalusa Heart Study; Bogaulsa, LA continued Bao, 199692 0.4-0.6 0.4-0.6 0.2-0.4 0.1-0.4 No Freedman, 198517 0.54-0.63 0.55-0.71 0.25-0.43 No, but values given Lipids appear to track better in are race and sex black M vs. black F or whites specific Berenson, 197993 0.42 0.43 0.14 0.21 Yes for year 1, race, Alphalipoprotein = HDL sex height at year 1, change in weight/height, change in ht. (MV regression) 0.58 0.69 0.23 0.31 No 0.49 0.64 0.30 0.35 No 0.52 0.58 0.41 0.22 No Child and Adolescent Trial for Cardiovascular Health Cohort Study; Multi-centered in U.S. continued Kelder, 20023 0.66 NR 0.62 NR 0.73 NR 0.73 NR 2 Evidence Table 1. Tracking of Serum Lipid Levels Study, year arranged by Duration of Fasting serum cohort N Demographics follow-up Age at start Age at follow-up levels? Child and Adolescent Trial for Cardiovascular Health Cohort Study; Multi-centered in U.S. continued Kelder, 20023 3,659 same 3 years 5th grade (mean 8th grade (mean No age 11.1 years) age 14.1 years) Muscatine Study; Muscatine, IA Clarke, 197894 7,371 3,521 M; 3,850 F 2 years 5-16 years 3,268 1,558 M; 1,710 F 4 years 6-14 years 816 408 M; 408 F 6 years 8-12 years Lauer, 198995 2,446 1,167 M; 1,279 F 12-14 years 8-18 years 20-30 years Yes (LRC method) 99 55 M; 54 F same 7-8 years 20-25 years 612 292 M; 311 F same 9-10 years 20-25 years 1,018 476 M; 542 F same 11-12 years 20-25 years 339 490 M; 551 F same 13-14 years 20-25 years 3 Evidence Table 1. Tracking of Serum Lipid Levels Correlation coefficients Study, year arranged by Other Adjusted? If so, cohort TC LDL HDL TG Outcome for what? Comments Child and Adolescent Trial for Cardiovascular Health Cohort Study; Multi-centered in U.S. continued Kelder, 20023 0.7 NR 0.65 NR No % remaining in highest quintile for 3rd grade to 5th grade: 59% for TC; 63% for HDL. % remaining in lowest quintile for 3rd grade to 5th grade: 59% for TC; 60% for HDL. For 3rd to 8th grade: lowest 53% for TC, 52% for HDL; highest 55% for TC, 55% for HDL. Muscatine Study; Muscatine, IA continued Clarke, 197894 0.68 0.44 No 0.63 0.4 No After 4 years, 40% of those in the highest quintile for TG remained in that quintile. 0.61 NA No After 6 years, 50% of those originally in the highest quintile for TC remained in that quintile. Lauer, 198995 Paper reports percent variability of adult lipids according to tobacco, alcohol and OCP use. M: 0.56, F: M: 0.56, F: M: 0.07, F: - NA No 0.64 0.65 0.25 M: 0.58, F: M: 0.56, F: M: 0.07, F: NA No 0.49 0.47 0.01 M: 0.51, F: M: 0.47, F: M: -0.04, F: NA No 0.54 0.53 0.06 M: 0.51, F: M: 0.47, F: M: -0.02, F: NA No 0.52 0.49 0.11 4 Evidence Table 1. Tracking of Serum Lipid Levels Study, year arranged by Duration of Fasting serum cohort N Demographics follow-up Age at start Age at follow-up levels? Muscatine Study; Muscatine, IA continued Lauer, 198995 1,041 155 M; 184 F same 13-14 years 26-30 years 767 352 M; 415 F same 15-16 years 20-25 years 568 263 M; 305 F same 15-16 years 26-30 years 615 299 M; 316 F same 17-18 years 20-25 years 479 233 M; 246 F same 17-18 years 26-30 years Beaver County Lipid Study, Beaver County, PA Stuhldreher, 295 48.7 M 16 years 11-14 years 27-30 (mean age Yes for follow-up 199196 28) values; baseline values measure calorimetrically by the ferric chloride-sulfuric acid method 5 Evidence Table 1. Tracking of Serum Lipid Levels Correlation coefficients Study, year arranged by Other Adjusted? If so, cohort TC LDL HDL TG Outcome for what? Comments Muscatine Study; Muscatine, IA continued Lauer, 198995 M: 0.52, F: M: 0.50, F: M: 0.16, F: NA No 0.52 0.49 0.08 M: 0.64, F: M: 0.60, F: M: -0.03, F: NA No 0.53 0.49 0.15 M: 0.63, F: M: 0.56, F: M: 0.06, F: NA No 0.55 0.50 0.10 M: 0.72, F: M: 0.63, F: M: -0.07, F: NA No 0.54 0.56 0.13 M: 0.64, F: M: 0.61, F: M: 0.06, F: NA No 0.48 0.45 0.04 Beaver County Lipid Study, Beaver County, PA continued Stuhldreher, M: 0.38; F: NA NA NA No 199196 0.51 6 Evidence Table 1. Tracking of Serum Lipid Levels Study, year arranged by Duration of Fasting serum cohort N Demographics follow-up Age at start Age at follow-up levels? Lipid Research Clinics Program, multi-centered, U.S. Namboodiri, 103 56 M, 47 F 0-30 months <20 years <20 years Yes, LRC 198497 (median 3 months) 105 55 M, 50 F 7-52 months <20 years <20 years Yes, LRC (median 29 months) 102 54 M, 48 F 7-102 months (not <20 years <20 years Yes, LRC explicitly stated) Laskarzewski, 108 46% M; 54% F; 81% 7 years 7-14 years 14-21 years Yes, LRC 197998 White, 18% Black, <1% Asian. Fels Longitudinal Study; Cincinnati, OH Guo, 199399 96 NR 4 years 9-21 years 13-25 Yes, LRC 96 NR 6 years 9-21 years 15-27 96 NR 8 years 9-21 years 17-29 96 NR 10 years 9-21 years 19-31 96 NR 12 years 9-21 years 21-33 Baumgartner, 69 27 F; 42 M unclear 12-18 years up to age 21 years Yes, LRC 1991100 100% White 7 Evidence Table 1. Tracking of Serum Lipid Levels Correlation coefficients Study, year arranged by Other Adjusted? If so, cohort TC LDL HDL TG Outcome for what? Comments Lipid Research Clinics Program, multi-centered, U.S. continued Namboodiri, M: 0.762, F: NA NA M: 0.374, F: Not reported, but 198497 0.782 0.510 paper states that "correlations were also computed with the advariate adjusted lipid values; the results were generally similar." M: 0.703, F: M: 0.672, F: M: 0.488, F: M: 0.436, M: As stated above 0.604 0.592 0.241 0.331 M: 0.593, F: NA NA M: 0.363, F: As stated above 0.604 0.522 Laskarzewski, 0.68 0.61 0.53 0.39 197998 Fels Longitudinal Study; Cincinnati, OH continued Guo, 199399 0.77 0.78 0.57 0.63 0.67 0.43 0.58 0.61 0.31 0.61 0.58 0.34 0.72 0.65 0.46 Baumgartner, M: 0.63, F: M: 0.60, F: M: 0.43, F: M: 0.58, F: Used Foulkes-Davis tracking 1991100 0.56 0.52 0.79 0.53 index (using Dallal program), TC, TG, LDL for boys and TC, HDL for girls were statistically significant. 8 Evidence Table 1. Tracking of Serum Lipid Levels Study, year arranged by Duration of Fasting serum cohort N Demographics follow-up Age at start Age at follow-up levels? Aerobics Center Longitudinal Study, Cooper Institute, Dallas Texas Eisenmann, 48 12 F; 36 M 11 yrs 12-18 years (mean mean age 26.6 Yes 2004110 Mean weigh 61.5kg age 15.8) NON-U.S. COHORTS Amsterdam Growth & Health Study Twisk, 1997101 181 14 years 13-16 years Yes Kindergarten Study, Basel, Switzerland Mohler, 1996102 NR NR 5 years 5 years 10 years No, capillary blood morning sample not fasting NR NR 4 years 10 years 14 years No 249 NR 9 years 5 years 14 years No Quebec, Canada Vobecky, 1988103 88 at risk children 41 M, 47 F 9 years 3-36 months 9-11 years Yes (LRC method) Quebec Family Study Katzmarzyk, 147 76 M, 71 F 12+1.5 years 8-18 years Yes 2001104 9 Evidence Table 1. Tracking of Serum Lipid Levels Correlation coefficients Study, year arranged by Other Adjusted? If so, cohort TC LDL HDL TG Outcome for what? Comments Aerobics Center Longitudinal Study, Cooper Institute, Dallas Texas continued Eisenmann, 0.62 NR 0.6 0.54 TC:HDL length of follow-up 2004110 ratio 0.78 used as a covariate NON-U.S. COHORTS continued Amsterdam Growth & Health Study continued Twisk, 1997101 tracking NA tracking NA coefficient: coefficient: M: 0.71 0.51, F: 0.65 Kindergarten Study, Basel, Switzerland continued Mohler, 1996102 0.07 (NS) 0.2 0.19 0.2 0.29 0.21 0.09 (NS) 0.19 0.16 Quebec, Canada continued Vobecky, 1988103 OR for OR for OR for TC>200 LDL>125 HDL<40mg/dL: mg/dL: 4.35 mg/dL: 3.88 2.10 (0.89- (1.78-10.63) (1.85-8.14) 4.97) Quebec Family Study continued Katzmarzyk, NA NA M: 0.58, F: M: 0.37, F: TC/HDL Yes, initial age and 2001104 0.56 0.20 ratio: M: length of follow-up 0.51, F: period 0.57 10 Evidence Table 1. Tracking of Serum Lipid Levels Study, year arranged by Duration of Fasting serum cohort N Demographics follow-up Age at start Age at follow-up levels? Helsinki, Finland Kallio, 1993105 162 NR 5 years birth 5 years Cord blood sample at birth, scalp vein sample at 2,4,6,9,12 mos, non-fasting 162 NR 2 months 5 years 162 NR 4 months 5 years 162 NR 6 months 5 years 162 NR 9 months 5 years 162 NR 12 months 5 years Toledo Area Study; Toldeo, Spain Bastida, 2002106 38 21 F, 17 M 4 years births 4 years 12 hour fast for 4 year olds, cord blood sample at birth Cardiovascular Risk in Young Finns Study; mulit-centered in Finland Porkka, 1991107 2,236 47.6% M 6 years 3-18 years Yes Porkka, 1994108 883 414 M, 469 F 12 years 3-18 years (15-30 years) Yes 187 97 M; 90 F 12 years 3 years 15 years Yes 160 74 M; 86 F 12 years 6 years 18 years Yes 11 Evidence Table 1. Tracking of Serum Lipid Levels Correlation coefficients Study, year arranged by Other Adjusted? If so, cohort TC LDL HDL TG Outcome for what? Comments Helsinki, Finland continued Kallio, 1993105 0.01 NA NA NA No 0.31 0.43 0.26 0.26 0.29 NR NR NR 0.32 0.2 0.21 0.27 0.36 0.41 0.39 0.52 0.47 0.58 0.29 0.37 No Toledo Area Study; Toldeo, Spain continued Bastida, 2002106 0.35 0.53 0.49 NR Cardiovascular Risk in Young Finns Study; mulit-centered in Finland continued Porkka, 1991107 0.63 0.66 0.58 0.36 No Porkka, 1994108 M: 0.58, F: M: 0.58, F: M: 0.58, F: M: 0.37, F: No M showed more tracking than F 0.48 0.53 0.53 0.33 M: 0.60, F: M: 0.62, F: M: 0.64, F: M: 0.25, F: No 0.51 0.57 0.51 0.46 M: 0.56, F: M: 0.60, F: M: 0.56, F: M: 0.13, F: No 0.37 0.47 0.44 0.32 12 Evidence Table 1. Tracking of Serum Lipid Levels Study, year arranged by Duration of Fasting serum cohort N Demographics follow-up Age at start Age at follow-up levels? Cardiovascular Risk in Young Finns Study; mulit-centered in Finland continued Porkka, 1994108 156 77 M; 79 F 12 years 9 years 21 years Yes 149 64 M; 85 F 12 years 12 years 24 years Yes 115 51 M; 64 F 12 years 15 years 27 years Yes 116 51 M; 65 F 12 years 18 years 30 years Yes Eno, East Finland Fuentes, 2003109 82 49 M, 33 F 8 years 7 years 15 years Yes Abbreviations M = Male; F = Female 13 Evidence Table 1. Tracking of Serum Lipid Levels Correlation coefficients Study, year arranged by Other Adjusted? If so, cohort TC LDL HDL TG Outcome for what? Comments Cardiovascular Risk in Young Finns Study; mulit-centered in Finland continued Porkka, 1994108 M: 0.63, F: M: 0.58, F: M: 0.56, F: M: 0.18, F: No 0.52 0.53 0.60 0.22 M: 0.57, F: M: 0.56, F: M: 0.69, F: M: 0.45, F: No 0.60 0.59 0.52 0.21 M: 0.40, F: M: 0.44, F: M: 0.57, F: M: 0.53, F: No 0.40 0.51 0.48 0.58 M: 0.73, F: M: 0.67, F: M: 0.54, F: M: 0.49, F: No 0.51 0.48 0.56 0.37 Eno, East Finland continued Fuentes, 2003109 Yes NA NA NA Gender, body TC decreased more steeply in weight, parental boys than in girls from age 7 to education, family 15, but no statistically significant history of obesity, age-gender interaction was family history high found. cholesterol, family history CVD and BMI Abbreviations M = Male; F = Female 14 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Andersen, 2003135 Randomly-selected 1020 To determine whether the Cardiovascular fitness: Estimated from maximal Danish children ages 9 number of participants with power output in a cycle test with progressively and 15 multiple coronary heart disease increasing workload until exhaustion. (CHD) risk factors exceeded the Anthropometric measures: SSK (biceps, triceps, number expected from a random subscapular and suprailiac) distribution. To establish whether BMI and to what extent clustering of Venous blood samples: Following overnight fast risk factors occurs in Danish Risk factors defined as: TC, HDL, TG, serum children and adolescents. insulin and blood pressure Bergstrom, 1997136 Healthy Swedish 879 To assess the relationship Activity: Self-report 7-day records and adolescents aged 14 to between physical fitness and BMI questionnaire. Physical training and other 17. and suggested cardiovascular physical exercise graded on a 6-graded scale. risk indicators of lipids, insulin, Cardiovascular fitness: 3 km running test ferritin and blood pressure; as Anthropometric: BMI, SSF (biceps, triceps, well as physical activity and subscapular, suprailiac skinfolds) dietary intake. Diet: Self-report 7-day record Blood: lipids, insulin, ferritin Blood pressure Chronic disease history Bonora, 1992137 18 yr old males, 1293 To evaluate the relation of either Body mass index (obese = BMI of >27) Verona Italy total body fat or body fat Waist and hip circumferences (body fat localization in several risk factors distribution) for atherosclerosis (serum lipids, Total body fat or fat free mass (electrical insulin, and blood pressure) resistance) Serum cholesterol (HDL, LDL, & Total) Triglycerides Insulin Blood pressure Smoking, alcohol use, physical activity 1 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Andersen, 2003135 X NE Those who had a high level of NE cardiovascular risk factors (4 or 5 risk factors) had a physical fitness level 1.2 standard deviations lower than mean values for the population. Authors conclude study found clustering of risk factors for metabolic syndrome. Bergstrom, 1997136 X Physical activity: + correlated Positive relationship between running Fat intake: + correlate HDL boys only HDL, boys only (0.008, 95%CI, time and TG (0.013, 95%CI, 0.003,0.023) 0.001,0.014) Boys: 0.014, 95%CI, 0.000,0.028 inverse correlate TG girls only Girls: 0.016, 95%CI, 0.005,0.027 (-0.013, 95%CI, -0.022,-0.003) Iron intake: Inverse correlate ferritin boys only (-37.995, 95%CI, -73.237,-2.752) Bonora, 1992137 XNENENE 2 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Andersen, 2003135 NE BMI: Those who had a high level of cardiovascular risk factors (4 or 5 risk factors) had a BMI 1.6 standard deviations higher than mean values for the population. Risk factors included TC, HDL, TG, serum insulin and blood pressure. Bergstrom, 1997136 NE BMI inversely related to HDL (boys -0.027, 95%CI, -0.044,- 0.009; girls -0.019, 95%CI,-0.031,-0.007) BMI + related to LDL (boys 0.037, 95%CI, 0.003,0.071; girls 0.036, 95%CI 0.006,0.067) BMI + related to insulin (boys 0.383, 95%CI, 0.193,0.572; girls 0.305, 95%CI, 0.128,0.483) Degree of adiposity is the most important factor in explaining differences between lipid and insulin values between adolescents with different levels of fitness. Bonora, 1992137 NE Total body fat is independently associated with: HDL, HDL/TC, TG, insulin, systolic blood pressure, diastolic blood pressure (p<.001). Not significant for total cholesterol or LDL. Total body fat is a better predictor of risk than waist/hip ratio (body fat localization). 3 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Andersen, 2003135 NE NE Bergstrom, 1997136 NE NE Bonora, 1992137 NE NE 4 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Bonora, 1996138 Healthy Italian men in 188 To examine to what extent Activity: Questionnaire on physical activity at work Verona, Italy aged 18 differences in total body fat and during leisure time, classification into light, or 38. Subset of those content, regional fat distribution, moderate and intense in the Verona Young and lifestyle contribute to Anthropometric: BMI, waist circumference, hip Men Artherosclerosis differences in the cardiovascular circumference, estimate of fat-free mass and total Risk Factors Study. risk profile of middle-age vs body fat (body electrical resistance) Our focus is on the 18 young men. Blood sample after overnight fast: TC, HDL, TG. year old population LDL calculated. Hypertriglyceridemia: TG > 250 only. mg/dL. Hypercholesterolemia: TC >240 mg/dL. HDL: abnormal if < 35 mg/dL. Insulin measured by Hales and Randle method. Smoking Daily alcohol intake: Grams consumed per day Blood pressure: Hypertension defined as systolic > 160 mmHg and/or diastolic > 95mmHg Demerath, 2003139 5th grade children in 14- 1338 To examine the feasibility and Anthropometric measures: BMI rural West Virginia utility of conducting a school- Blood samples: Whole blood from finger prick, counties based obesity and cardiovascular with portable Cholestech LDX analyzers to disease risk factor screening measure TC and HDL-C. program in a rural Appalachian population. DeStefano, 1995140 Schoolchildren in rural 2726 To evaluate determinants, Family history: Brief questionnaire completed by Wisconsin aged 5 to 15 for multi- especially dietary, of serum lipid parents or children variate concentrations in schoolchildren Anthropometric measures: BMI analysis in a community-wide screening. Diet: brief questionnaire completed by parents or children. Most diet questions open-ended with questions on consumption during past week or during average day. Venous blood samples after 12 hour overnight fast: TC, HDL, calculated LDL 5 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Bonora, 1996138 X Not included in multiple linear Not included in multiple linear NE regression analysis regression analysis Demerath, 2003139 XNENENE DeStefano, 1995140 X NE NE TC and LDL: Associated with butter and egg consumption HDL: Associated with butter and lard, negative association with vegetable shortening, whole grain bread Triglycerides: Associated with magerine, nonfat milk; negative association with eggs. TC:HDL ratio: no strong association with diet 6 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Bonora, 1996138 NE Only muliple linear regression analysis results reported here. BMI, waist/hip ratio, and insulin levels strongly intercorrelated. BMI predicted TG (30% of this risk factor variability). Std coefficient 0.48, t-value 3.1, p<0.005. Demerath, 2003139 NE TC: Greater risk of high TC among obese vs non-overweight (OR 2.4, p< 0.05), sensitivity 50%, specificity 71% HDL: Lower among obese vs non-overweight (OR 5.3, p<0.05), sensitivity 66%, specificity 73% DeStefano, 1995140 TC and LDL: strong association with family history of Increased BMI: Related to adverse levels of all lipids and hypercholesterolemia (r=0.184 and 0.174, respectively) lipoproteins (TC, LDL, HDL, TG). TC:HDL ratio: associated with having 1st degree relative with BMI: Associated with TC:HDL ratio hypercholesterolemia. 7 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Bonora, 1996138 NE NE Demerath, 2003139 NE NE DeStefano, 1995140 NE Parent who smokes: negative association with TC:HDL ratio 8 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Douglas, 1996141 Youth aged 5 - 19, 506 To determine whether adding the Family history: Data collected at appointment. predominantly African criterion of obesity to family Anthropometric: BMI American and history would increase the yield Diet: Question from HANES-1 Hispanic, who live in of screening Blood samples: Nonfasting cholesterol (TC, TG) an inner city via finger stick. Those with elevated TC were environment. asked to return for a fasting lipid profile via venipuncture (TC, TG, HDL). Elevated TC: 170 mg/dl Elevated TG: 100 mg/dl 9 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Douglas, 1996141 XNENENE 10 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Douglas, 1996141 Predictor of higher cholesterol vs no family history (p<0.001). Obesity defined as BMI > 20% above ideal BMI. Odds ratio of those with family history of hypercholesterolemia Predictor of higher cholesterol: correlation to non-fasting TC: having high cholesterol: 2.28 (95% CI=1.70 to 2.86). p<0.001. Sensitivity of family history as screening tool: 24% OR of obese screenees having hypercholesterolemia: 1.74 Use of family history of CVD, hypertension, myocardial (95% CI=1.34 - 2.14). infarction, stroke produced similar results. Multiple linear regression analysis: obesity related to Specificity: 88% hypercholesterolemia when controlling for age and family Positive predictive value: 51% history (p<0.001). Negative predictive value: 69% Sensitivity in screening for hypercholesterolemia: 42% Specificity: 71% Positive predictive value: 41% Negative predictive value: 72% Predictive value of family history and obesity combined: Sensitivity: 49%, Specificity: 64%, Positive predictive value: 41%, Negative predictive value: 71% 11 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Douglas, 1996141 NE NE 12 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used DuRant, 1982142 Black children and 84 To evaluate the influence of a Family history: Family health questionnaire re: 1st adolescents aged 7 - family history of CHD and CHD and 2nd degree relatives administered to parent. 15 who were risk factors on the TC/HDL and Anthropometric measures: Triceps skinfold economically LDL/HDL ratios in a group of thickness disadvantaged and black children and adolescents Diet: 24-hour intake history participated in a 6- Venous blood samples: Drawn after 12-hour fast, week organized TC, TG, HDL. LDL calculated. community recreation program 13 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet DuRant, 1982142 XNENENE 14 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures DuRant, 1982142 Parental CHD risk factors with significant effects on LDL/HDL NE and TC/HDL: stroke, diabetes mellitus, obesity. Stroke and age, diabetes and age, and obesity and age interactions in TC/HDL, LDL/HDL. Children with family history of stroke: 15 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous DuRant, 1982142 NE NE 16 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used DuRant, 1983143 Middle class white 50 To examine the relationship Activity: Questionnaire of usual level of physical males aged 11 -17 between reported physical activity (hours watching tv, reading and studying, recruited from local boy activity levels and HDL, LDL team sports during previous school year, scout troops. /HDL ratio, and TC/LDL among sports/activities performed > 3 days/week white male adolescents. continuously for > 1 hour at a time, days of jogging per week, days bike riding per week) Exercise capacity tested by mechanically braked bicycle ergometer. Anthropometric: Weight, height, relative body weight (Weight-for-height index) triceps skinfold, resting pulse, blood pressure Diet: 48-hour diet history kept by parents, with dietary recall completed by a registered dietician. Blood sample after 12 hour fast: TC, HDL, LDL Other: Cigarettes smoked per day, alcohol consumed per week. DuRant, 1993144 4 and 5 year old 123 To examine relationships among NA African American, indicators of physical activity, Hispanic and physical fitness, and body Caucasian children in composition with serum lipid and the Texas Studies of lipoprotein levels in young Child Activity and children. Nutrition (SCAN) program. 17 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet DuRant, 1983143 X Days jogged per week: 2nd Maximum exercise capacity not Not significant in multiple regression strongest predictor of TC/HDL related to lipid levels. (r=0.29); strongest predictor of LDL/HDL (r=-0.32) DuRant, 1993144 X TC: Not significant TG: negative correlation with NE LDL: Not significant percentage of time in higher activity TG: Negative correlation with level, work load. mean activity level 18 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures DuRant, 1983143 NE NE DuRant, 1993144 NE TG: positive correlation with sum of skinfolds at year 1 and 2, and with waist/hip ratio and heart rate at year 2. TSC: at year 2, correlated with year 1 waist/hip ratio, but year 2 waist/hip ratio not significant LDL: correlated with year 1 waist/hip ratio HDL: correlated with sum of skinfolds at year 2 LDL/HDL ratio: positive correlation with year 1 waist/hip and year 2 sum of skinfolds TSC/HDL ratio: positive correlation with year 1 waist/hip and year 2 sum of skinfolds 19 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous DuRant, 1983143 Age: Strongest predictor of HDL (r=-0.42), and TC/HDL Television watched per night: 2nd strongest predictor of HDL- (r=0.38) (r=-0.42), 3rd strongest predictor of TC/HDL Height: 3rd strongest predictor of HDL (r=-0.31) Hours spent reading per day: 2nd strongest predictor of Systolic blood pressure: predictor of TC/HDL (r=0.30); of LDL/HDL (r=0.27) LDL/HDL (r=0.23) Smoking and alcohol use not significant in multiple regression. DuRant, 1993144 NE NE 20 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Dwyer, 1994145 Australian 1919 To relate objective measures of Endurance fitness: physical work capacity on a schoolchildren aged 7 - fitness to standard CHD risk Monark bicycle ergometer. 15 factors on a large representative Measure for fitness analysis: ratio of work sample of children, to assess capacity to lean body mass. overall associations and Anthropometric: BMI, skinfold thickness at 4 sites determine whether threshold Serum lipids: measured after 12 hour fast. Lipid effects are present. Research Clinic procedures followed. Plasma total cholesterol, triglycerides and HDL determined. Eisenmann, 2002146 Runners between age 54 To examine age- and sex- Activity: Self-reported running volumes 8 and 15 who associated variation in Anthropometric: SUM6: subcutaneous measures consistently placed subcutaneaous andipose tissue of triceps, biceps, subscapular, supra-iliac, within the top 5 (SAT) and its association with abdominal and medial calf. TER: (trunk-to- finishers of road races blood lipoproteins among extremity ratio) ratio of 3 trunk skinfolds to sum of 10 km or more by age adolescent distance runners. 3 extremity skinfolds. SAT: (subcutaneous and gender. adipose tissue) TER regressed on SUM6, residuals retained to represent an index of relative SAT independent of overall subcutaneous fatness Venous blood samples: following 12 hour fast and 24 hours after last bout of exercise. TC, TG, HDL- C, LDL-C. 21 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Dwyer, 1994145 X Not significant for TC or HDL. Not significant for TC or HDL. NE Eisenmann, 2002146 X All subjects very active: self- NE NE reported mean training volumes 1503 (males) and 1865 (females) km per year . 22 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Dwyer, 1994145 NE BMI TC - Negative correlation (p<.01) HDL - Negative correlation (p<.001) TG - correlated (p<.001) Sum of skinfolds: TG - correlated (p<.001) HDL - correlated (p<.001) Eisenmann, 2002146 NE Increase in SUM6: associated with > in LDL-C, TG and TC:HDL in adolescent males (p<0.05) In adolescent females: associated with < in HDL-C and > in TG (p<0.05). TER not significant. 23 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Dwyer, 1994145 NE NE Eisenmann, 2002146 NE NE 24 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Freedman, 199918 5 - 17 year old males 9167 To examine the relationship of Anthropometric: Quetlet index/BMI (kg/m2). and females examined cut points for overweight derived Triceps and subscapular skinfolds. Overweight: in 7 cross-sectional from several national studies defined as Quetlet index above the smoothed studies conducted by (Quetelet index, >95th percentile) 95th percentile of combined data from 5 national the Bogalusa Heart to adverse risk factor levels and surveys conducted between 1963 and 1994. Study between 1973 risk factor clustering. and 1994. Blood samples: TC, TG, LDL, HDL. High level of TC: >200 mg/dL, TG: > 130 mg/dL, LDL: 130 mg/dL. HDL: <35 mg/dL considered low. Fripp, 1985147 Males aged 15 -17 37 To evaluate the interrelationship Family history questionnaire. Aerobic capacity: recruited from the 10th between aerobic capacity, progressive treadmill test. Anthropometric: BMI, grade in a regional ponderosity, and atherosclerotic ponderal index. Venous blood samples after public high school who risk factors by examining overnight fast: TC, TG, LDL-C, HDL-C, LDL-B. tested as having low- exercise duration, maximum moderate to moderate oxygen consumption, body mass levels of physical index, systolic and diastolic blood fitness pressure, and levels of plasma lipids of male adolescents with low-to-moderate levels of physical fitness. 25 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Freedman, 199918 XNENENE Fripp, 1985147 X NE HDL-C higher with increased NE exercise duration (p<.05) Triglyceride level lower with increased exercise duration (p<.05). 26 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Freedman, 199918 NE Overweight: defined as Quetlet index above the smoothed 95th percentile of combined data from 5 national surveys conducted between 1963 and 1994. TC: Overweight children 2.4x more likely to have elevated TC LDL-C: 3.4 x more likely TG: 7.1 x more likely HDL-C: 3.4x more likely to be <35mg/dL Use of overweight as screening tool could identify 50% of school children who had 2 or more risk factors. Sensitivity for TC: 24% TG: 47% LDL: 28% HDL: 25% Fripp, 1985147 NE BMI: positive association with triglyceride level (p<.05); negative association with HDL-C (p<.05). 27 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Freedman, 199918 NE NE Fripp, 1985147 NE NE 28 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Giovannini, 1992148 Schoolchildren in 2 361 To investigate how levels of Family history: Parent report of 1st and 2nd small towns near Milan, pediatric lipoprotein are degree history of atherosclerotic complications. Italy influenced by nutritional and Anthropometric: weight > 20% of ideal labeled as familial factors in a northern obese Italian population. 3-day diet record Blood samples: TC, HDL-C, TG, ApoB, ApoAI. LDL calculated by Friedwald's formula. Glassman, 1993149 Retrospective chart 98 To determine whether inclusion Family history: obtained from medical record review of 98 children of obesity to the current AAP Anthropometric measures: BMI, percent ideal referred for evaluation criteria significantly enhances body weight [(actual weight/height appropriate and treatment of their predictive value in weight)x100]. Obesity defined as BMI 20.5 or obesity or Type IIA identifying children with body weight >120% of ideal. hyperlipoproteinemia at hypercholesterolemia. Venous blood samples after 14 hour fast: TC, the Children's Nutrition TG, HDL-C. LDL-C calculated as difference Center, NY Medical between TC and HDL-C. College. 29 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Giovannini, 1992148 X NE NE Higher TC/HDL-C ratio in the lower quartile of polyunsaturated fatty acid intake (survey data, no p value reported) Glassman, 1993149 XNENENE 30 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Giovannini, 1992148 CVD not significant for TC, LDL TC: higher for obese girls (p<0.01) Hypercholesterolemic parents: associated with higher TC and TG: higher for obese girls and boys (p<0.01) LDL, although not predictive of TC>200 mg/dl or LDL >120 HDL-C: lower for obese girls (p<0.001) mg/dl. LDL-C: higher for obese girls (p<0.001) Family history of CVD: ApoB: higher for obese girls (p<0.001) and boys (p<0.1) Sensitivity TC>200: 0.19, LDL>140: 0.09 ApoA1: not significant Specificity: 0.83, 0.91 LDL-C/ApoB: lower in obese boys + predictive value: 0.34, 0.30 (p<0.1) - predictive value: 0.69, 0.69 of Family history of HC: Sensitivity 0.22 0.13 Specificity 0.85 0.93 + predictive value: 0.47 0.52 - predictive value: 0.65 0.65 Family history of CVD + HC: Sensitivity 0.20 0.11 Specificity 0.86 0.93 + predictive value: 0.61 0.64 - predictive value: 0.48 0.48 Glassman, 1993149 NE Obesity defined as BMI 20.5 or body weight >120% of ideal. 36/45 (80%) obese children had cholesterol levels >90th percentile (suggesting > risk for hypercholesterolemia in this group). If obesity was added to AAP criteria, 66/80 hypercholesterolemic subjects would have been identified. This would also improve sensitivity of the standards (70 vs 87%, p<0.02) and negative predictive value (45 vs 30%, p<0.02) BMI and percent ideal weight not correlated with TC/LDL-C in obese subjects. BMI inversely correlated with TC (p<0.005) and LDL-C (p<0.002) in non-obese, hypercholesterolemic patients. 31 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Giovannini, 1992148 NE NE Glassman, 1993149 NE NE 32 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Gliksman, 1993174 Healthy Australian 1017 To report on differences in Anthropometric: BMI schoolchildren aged 12 dietary variables and serum lipids Diet: 24 hour diet record and 15 (subgroup of by age, sex, and ethnic origin in Blood sampling after 12 hour fast with procedures 1985 ACHPER sample a national cohort of Australian from LRC. TC, TG, HDL. LDL calculated. survey) children. Socioeconomic status: As linked to postal code. Ethnic origin: Determined by parent birth country Howard, 1991150 Children aged 7-18 78 To answer 6 research questions: Bloomsday Cardiovascular Fitness years in Spokane, WA Is there a difference in: Questionnaire; Coronary Risk Profile (family 1) Physical measurements and history and other demographics); Diet Habit different diets Survey: Type of diet: 1) meat 2) dairy 3) fats and 2) Physical measurements and oil 4) grains 5) sweets 6) salt Four classifications blood pressures of fat intake: American diet (40% from fat), Phase 3) Physical measurements and I (30% fat), II (25% fat), and III (20% fat). parental smoking status Physical measurements: total serum cholesterol 4) Physical measurements and and HDL-C level, weight, and blood pressure family history of CD or no CD 5) Relationships exercise and physical measurements 6) Physical measurements and Type A personality 33 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Gliksman, 1993174 X NE NE In multiple regression analysis, socioeconomic and gender-based lipid differences could be explained by differences in diet. Significant differences of dietary fat intake based on ethnic origin were not reflected in serum lipid differences. Girls: Dietary fat variables were more important predictors of serum lipids than BMI. (%kJ saturated fat: TG, HDL; PS ratio: TG, HDL, LDL/HDL) Boys: BMI was a more important predictor of serum lipids than dietary fat.(Dietary fat not associated with TC, TG, HDL, LDL, LDL/HDL) Howard, 1991150 Multiple one Walk and run: Pearson Correlation only American diet of meat had significantly way ANOVAs correlations only lower weight percentiles than Phase II diets (p =0.03). American diet of fat had significantly higher HDL-C percentiles than those in Phase I, II, or III (p=0.001). American grain diet (>13 yrs) had higher cholesterol percentiles than subjects in Phase II (p=0.01). Phase I diet (>13 yrs) higher cholesterol than subjects in Phase II (p=0.01). American grain diet lower weight percentiles than Phase 1 (p=0.01). 34 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Gliksman, 1993174 NE Girls: Dietary fat variables were more important predictors of serum lipids than BMI. BMI negatively associated with HDL. Positive association with TG and LDL/HDL. Boys: BMI was a more important predictor of serum lipids than dietary fat. TC and HDL: Negative association with BMI. TG and LDL/HDL: Positive association with BMI. Howard, 1991150 Family history of heart disease and high blood pressure: Chi NE square tests only 35 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Gliksman, 1993174 In multiple regression analysis, age did not explain NE socioeconomic and gender-based differences in serum lipids. Howard, 1991150 Blood pressure: Mann-Whitney U statistics only Stress/personality characteristics: No need to excel had higher weight percentiles than those with some, a fair or a great need to excel (p=0.04, p=0.01, p=0.001 respectively). Non- competitive subjects had a higher weight percentile than those who were somewhat competitive or very competitive (p=.0.05, p=0.001 respectively). Smoking: T-tests only 36 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Jiang, 1995173 Bogalusa, LA, 5-30 yr 4136 To assess whether circulating Serum total cholesterol, triglycerides, VLDL, HDL, olds (age 5 - 17 insulin is a major contributor to LDL, plasma glucose, plasma insulin. reported here ) adverse lipid profiles during the Obesity (skinfold thickness), smoking, alcohol transition from adolescence to and life style and health habits survey adulthood Jarvisalo, 2001151 Males and females 88 To examine the feasibility of Abdominal aorta (aIMT) and carotid artery ages 8-14 yrs measuring aIMT in children and thickness (cIMT) ; blood pressure; serum total, Non-controls from to study its value in distinguishing LDL and HDL cholesterol; triglycerides; glucose outpatient clinic of the high-risk children from healthy Department of controls compared with a more Pediatrics, Turku established marker of subclinical University Central atherosclerosis Kunz, 2005153 Boys with mean age 336 To compare coronary risk factors TG, HDL-C, TC, gluclose, LDL-C (Friedman's 14.5 recruited from 8 of smoking and non-smoking formula), coagulometric and chromogenic assays, schools in Innsbruck, boys; to gather information on radioimmunoassays, blood cell counts. Austria. whether boys who start smoking Cardiovascular endurance (PWC-170 bicycle early have different baseline ergometer test), handgrip strength, average charactoristics impacting weekly hours in sports, forced expiratory vital coronary risk. capacity and peak flow. Self report of medical histoy, sports activities, alcohol consumption, smoking habits of self and parents. Kwiterovich, 1997154 Boys and girls with 663 To examine the effects of dietary Fasting serum LDL-C, total cholesterol, elevated LDL age 8 - intake of fat and cholesterol, triglycerides, HDL-C, dietary intake, height, 10 at baseline of 3 year sexual maturation and BMI on weight, Tanner stage. study LDL-C on boys and girls with elevated LDL-C levels. 37 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Jiang, 1995173 XNENENE Jarvisalo, 2001151 XNENENE Kunz, 2005153 XNENENE Kwiterovich, 1997154 X NE NE For boys, LDL-C decreased by 0.696 mg/dL for each 10 mg/4.2 MJ decrease in dietary cholesterol (p<.05). For girls, no single nutrient was significant, but there was a treatment group effect (p<.05) 38 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Jiang, 1995173 NE NE Jarvisalo, 2001151 NE NE Kunz, 2005153 NE NE Kwiterovich, 1997154 NE BMI For both genders, BMI at 3 years and LDL at baseline were significant and positive predictors of LDL 39 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Jiang, 1995173 Fasting insulin level: + association with VLDL cholesterol NE independent of age, sex, glucose concentration, obesity, cigarette smoking, and alcohol intake; - association with HDL in whites aged 5 to 17. Although fasting insulin level along with other variables was shown to be a significant predictor of lipoprotein measures, it can account for only a small part of the variability of lipoprotein cholesterol level (R2 values were low). Jarvisalo, 2001151 Multivariate regression model results: NE aIMT associated with group (diabetes and HC) (p=0.001), age (p=0.014), diastolic blood pressure (p=0.75). cIMT was associated with group (diabetes and HC) (p=.0.001) and diastolic blood pressure (p=0.046). No association with serum total, LDL and HDL cholesterol, triglycerides or glucose Kunz, 2005153 NE Smoking TC and LDL-C lower in smokers after adjustment for age, size, weight, BMI. HDL-C: No significant difference between smokers and non- smokers after adjustment for weight or BMI TG: No difference Kwiterovich, 1997154 NE Sexual development Males: LDL lower at Tanner Stage 4+ than Stage 1 (p<.01) Females: LDL lower at Tanner Stage 4+ than Stage 1 (p<.05) 40 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Larsson, 1992152 7 year old students in 147 with To compare lipid measures in 7 Family history: Parents completed a family history Sweden family year old children who have a questionnaire history; family history of myocardial Venous blood samples: Drawn after overnight 148 infarction/angina pectoris before fast, TC, HDL-C, TG. LDL calculated. controls age 55 or hyperlipidemia to children without this family history. Macek, 1989155 Students aged 16 - 18. 93 To investigate differences in Family history: Parents completed World Health Treatment group CHD risk factors in groups of Organization Questionnaire for CHD history. recruited from high trained and untrained Maximal aerobic power: cycle ergometer following school where 5 hours a adolescents. WHO criteria. Anthropometric: percentage body day was devoted to fat. Venous blood samples following overnight swimming and fast: TC, HDL-C, LDL-C, Apo-A, Apo-A1, Apo-B, athletics. Controls, TG. recruited from a similar school, had lower activity levels, were 1.7 years older and similar in height and therefore viewed as comparable. Marti, 1989156 15 year-old boys and 1142 To examine the associations Leisure time exercise: self-report using a 5 girls from 40 Finnish between frequency of leisure category question. schools time exercise and cardiovascular Aerobic fitness: Self-report of distance covered risk factors in adolescents. during 12 minute run. Anthropometric: BMI, Tanner scale for sexual maturity Blood samples: serum TC, HDL, serum thiocyanate. Smoking measured by self-report in 13 questions 41 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Larsson, 1992152 XNENENE Macek, 1989155 X NE TC: not significant NE HDL-C higher vs controls (p<0.05); significant relationship with Vo2 max (p<0.001) Apo-A1 higher vs controls (p<0.01) Apo-A higher vs controls (p<0.01) LDL-C lower vs controls (p<0.001) Apo-B lower vs controls (P<0.001) TG lower vs controls (p<0.001); significant relationship with Vo2 max (p<0.001) Marti, 1989156 X No relationship between leisure No relationship between aerobic NE time exercise and TC or HDL. fitness and HDL or TC. 42 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Larsson, 1992152 TC, LDL, LDL/HDL for those with family history of NE hyperlipidemia vs controls (p<0.001). No difference in these lipids among those with family history for myocardial infarction or angina pectoris exclusively vs controls. HDL, TG: No significant difference between those with any family history vs controls Macek, 1989155 NE NE Marti, 1989156 NE HDL: Inversely related to BMI for boys and girls (p<0.01) and sexual maturation for boys (p<0.01). 43 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Larsson, 1992152 NE NE Macek, 1989155 Gender: Girls had higher HDL-C (p<.001) NE Marti, 1989156 NE Smoking: Inversely related to HDL for boys (p<0.01) and girls (p<0.05) 44 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Muhonen, 1994120 9th through 12th grade 599 To determine the utility of a Family history: HFTQ questionnaire, history of students aged 14 to 20 school-based family history heart attack, angina, bypass surgery, in Muscatine, LA. questionnaire to identify hypertension and obesity; morbidity and mortality adolescents with adverse in parents and grandparents coronary risk factor levels. Anthropometric measures: TSF thickness Venous blood samples after overnight fast: TC, HDL, TG. LDL calculated Raitakari, 1996157 Finnish children and 714 To analyze the effect of changes Physical activity index: questionnaire of leisure- adolescents aged 3-18 in physical activity and other time physical activity using intensity X estimated years, pulled from the determinants of HDL-C on HDL- duration X monthly frequency, the range was 0- national population C levels. 225. register Anthropometric measures: subscapular skinfolds, height, and BMI. Smoking: self-report of smoking status and frequency. Blood samples: serum total cholesterol, triglycerides, high-density cholesterol, and insulin. Resnicow, 1993158 Data obtained from 11,370 To examine the relationship Anthropometric: Quetelet's BMI (kg/m2) Know Your Body between BMI, systolic blood Blood samples: Non-fasting finger-stick. TC. High school health pressure (SBP), and total plasma TC defined as 200mg/dL promotion program cholesterol along with the 1984 - 1991, including implications of using obesity as a 29 elementary schools criterion for blood pressure in 8 states. screening in an ethically diverse Convenience sample; sample of 11,370 schoolchildren. half of schools from low-income, inner-city neighborhoods. 45 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Muhonen, 1994120 XNENENE Raitakari, 1996157 X Boys: HDL-C was significantly NE NE correlated with physical activity, with a correlation coefficient of 0.20 (p<0.001) Girls: Physical activity did not significantly correlate with HDL- C. Resnicow, 1993158 XNENENE 46 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Muhonen, 1994120 Parent history of CHD: >BMI, Raitakari, 1996157 NE Changes in subscapular skinfold thickness correlated to changes in HDL-C for both boys and girls (p<0.05 for both). BMI change was correlated with a change in HDL-C for girls only (p<0.001), BMI cut-off not defined. Resnicow, 1993158 NE 26% of subjects with BMI's in the top decile and elevated SBP had TC levels > 200 mg/dL. This was approximately twice the rate of high TC observed among overweight children with normal SBP (14%), normal-weight children with normal SBP (11%), and normal-weight children with elevated SBP (13%). 47 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Muhonen, 1994120 NE NE Raitakari, 1996157 NE NE Resnicow, 1993158 NE NE 48 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Ribeiro, 2004159 Healthy, treatment-free 1533 1) To determine the coexistence Activity: Physical activity index (PAI) self-report children aged 8 to 15 in of CVD risk factors high blood questionnaire on activities > 15 minute duration in 17 primary and 13 high pressure (HBP), percentage of past week. Intensity catagories based on schools an urban and high fat body mass (%HFM), and metabolic equivalents. Risk factor: < 25th rural area of Porto, high total cholesterol (HTC) percentile of PAI adjusted for age & sex. PAI Portugal. 2) To study the relationship divided into quartiles. between physical activity (PA) Anthropometric: BMI, tricipital skinfold, and biological risk factors subscapular skinfold, percentage fat mass (%FM) clustering (HBP, %HFM, HTC). estimated via Slaughter, et al (skinfold and maturational stage measurements). Risk factor HFM: above 75th percentile for age and sex. Blood: Capillary samples from earlobe after > 12 hour fast for plasma TC. High TC (HTC): > 75th percentile adjusted for age, sex. Blood pressure: Systolic and diastolic measured in right arm. Risk factor HBP: SBP or DBP > 75th percentile adjusted for age, sex. Shear, 1985160 Children aged 5 to 17 3312 To assess the relationship Family history: Parent or guardian self-report on residing in Bogalusa, between parental history of questionnaire LA, and their biological vascular disease (heart attack, Venous blood samples: 12 hour fasting. TC, TG, parents. Part of the 4th stroke, diabetes mellitus and LDL, VLDL, HDL cross-sectional survey hypertension) and risk factors of the Bogalusa variables for cardiovascular prospective cohort disease in their offspring. study. 49 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Ribeiro, 2004159 X PAI and percent of subjects NE NE with biological risk factors Male: PAI 1st quartile, 59.1% had risk factors, 4th quartile, 41.3% had risk factors (p=0.010) Female: Not significant. PAI quartiles and 1 and 2 - 3 biological risk factors: no statistical significance. Shear, 1985160 XNENENE 50 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Ribeiro, 2004159 NE Percentile high fat mass and HBP: Highest observed odds ratio for clustering. Females: OR 2.6, CI 1.8-3.8 Males: OR 1.9, CI 1.3-2.8. Percentile high fat mass and high TC: No significant clustering. Shear, 1985160 Multivariate analysis: No significant effect for a single parental NE vascular disease. Combination of parental heart attack with parental diabetes (p<0.0001) and parental heart attack with parental hypertension (p<0.01) significant in relationship to overall risk factors in children. Risk factors: TC, TG, LDL, HDL, blood pressure 51 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Ribeiro, 2004159 About half of subjects had > 1 biological risk factor. NE HTC and HBP: Clustering significant in females (OR 1.6, CI 1.0-2.4) Shear, 1985160 NE NE 52 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Simon, 1995161 Premenarchal Black 1397 To determine the correlates of Activity: Television viewing per week and white girls aged 9 serum HDL-C in 9 and 10 year Anthropometric: BMI. Sum of skinfolds (triceps, and 10 in Richmond, old girls. subscapular, suprailiac). Tanner staging for CA; Cincinnati, OH; sexual maturation. and Washington, DC Diet: 3-day diet record Blood sample: HDL Suter, 1993162 Boys and girls aged 10 - 97 1) To describe lipid levels, Activity: 7-day recall with metabolic equivalents 15 recruited from a physical activity, fitness and diet (MET) and additive 7-day physical activity index private school in of healthy Canadians aged 10 - (PAIND) calculated. Calgary. 15. 2) To evaluate if predictive Substance use and oral contraceptives: assessed power of physical activity for in private interview with trained counselor. serum lipid levels exists for this Cardiovascular fitness: submaximal test using population as it does in adults. 3) bicycle ergometer. Anthropometric: obtained by 1 To adjust the relationship measurer with high reliability; sum of skinfolds between activity and blood lipids and BMI included. Diet: analysis of 3-day food for confounding factors. records. Venous blood samples after overnight fast: TC, TG, VLDL-C, LDL-C, HDL-C, Apo A-1, Apo B, lipoprotein (a). 53 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Simon, 1995161 X Not significant after multivariate NE Ratio of polyunsaturated fat/saturated fat adjustment for television intake and polyunsaturated viewing and pubertal fat/monosaturated fat intake as well as maturation. total fat intake showed no significant association with HDL in multivariate analysis. Polyunsaturated fat intake associated with higher HDL (p<.05). Each 10% > in polyunsaturated fat intake was associated with > of 3.4 mg/dL. Suter, 1993162 X Multiple linear regression Cardiovascular fitness not significant Multiple linear regression analysis: dietary analysis: significant predictor for TG, HDL, VLDL. fat intake not a predictor of TG, HDL, for HDL-C, Apo A-1, TG and VLDL. VLDL-C. 54 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Simon, 1995161 NE Changes in subscapular skinfold thickness correlated to changes in HDL-C for both boys and girls (p<0.05 for both). BMI change was correlated with a change in HDL-C for girls only (p<0.001), BMI cut-off not defined. Suter, 1993162 NE Multiple linear regression analysis: BMI not significant predictors of TG, LDL, VLDL. Gender is a modifying factor of the relationship between subcutaneous body fat and TG or VLDL-C levels. Bivariate analysis: High sum of skinfolds more closely related to an unfavorable lipid profile in girls than boys. 55 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Simon, 1995161 NE NE Suter, 1993162 NE Alcohol: 7/97 reported use, all in very small amounts. No statistical significance. Smoking: No subjects smoked. Oral contraceptives: No subjects used. 56 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Thorland, 1981163 8 - 11 year old boys 55 Compare lipid levels of pre- Activity: 5-day records with time-motion analysis. recruited from adolescent boys who have high Parents observed or got information from children elementary schools or low intensity of physical activity on activities and intensity. and sports clubs in Ann and high or low levels of body Anthropometric: Skinfold measurements at 6 Arbor, MI. fatness. sites, underwater weighing to assess relative fatness, fat weight, and lean body weight. Diet: 5 day record. Parents recorded food items and amounts consumed by child each day. Consumption away from home based on child report. Venous blood samples after overnight fast and 36 - 40 hours after exercise: TC, TG Tolfey, 1999164 Healthy 10 year old 71 To identify independent Activity: estimated through 4-day continuous heart boys and girls in contributions from predictor rate measurement of cardiovascular stress. England variables of habitual physical PAHR-25 (Total time spent 25% above resting activity, peakV02, percent body heart rate), PAHR-50 (50% above heart rate) and fat, and dietary composition are mean heart rate over collection period. related to prepubertal children's Cardiovascular fitness: Peak VO2 measured from lipid-lipoprotein profile. cycle ergometer test to volitional exhaustion with expired air collection. Anthropometric: Skin fold, body mass, estimated percent body fat. Sexual maturity assessed by parent and child responses to Tanner drawings. Diet: 7-day diary completed by parent Venous blood samples: Following 11-12 hour fast and no structured exercise 48 hours prior to sampling. TC, TG, HDL-C, , VLDL-C, LDL-C, TC/HDL-C and LDL-C/HDL-C derived from concentration. NCEP performance criteria met. 57 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Thorland, 1981163 X HDL/TC ratio larger in more NE Not significant 2-way active vs less active boys ANOVA and (p<0.05). stratification Triglyceride lower in more active boys (p<0.05). Tolfey, 1999164 X Girls: Habitual physical activity Girls: VO2 related to lipid-lipoprotein Girls: Polyunsaturated-saturated fatty acid (HPA) related to lipid- profile (p<0.05) in bivariate analysis. ratio accounted for 10.3% of LDL-C lipoprotein profile (p<0.05) Multivariate analysis: variance and (Bivariate analysis). 1 of 3 main in TG: VO2 accounted for 22.7% of 7.9% of TC/HDL-C variance predictor variables for girls in unique variance multiple regression analysis, in HDL-C: accounted for 24.8% of Boys: TC: Daily energy intake accounted accounting for 18% of unique unique variance for 15.4% of variance and variance in TC and 22.6% in in LDL-C/HDL-C: accounted for 22% of LDL variance LDL. 22.5% of unique variance in TC/HDL-C: accounted for 24.2% of Boys: HPA related to TC/HDL- unique variance C and LDL-C/HDL-C (p<0.05 for both) in bivariate analysis. Boys: Not significant Not significant in multiple linear regression. 58 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Thorland, 1981163 NE Body fatness not significant. Tolfey, 1999164 NE Multivariate analysis: TC: % body fat accounted for 12.1% of unique variance, girls only LDL: % body fat accounted for 21.2% of unique variance, girls only Boys: % body fat not significant 59 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Thorland, 1981163 NE NE Tolfey, 1999164 NE TG: Higher in girls (p<0.01) than boys, and the only gender difference in lipid-lipoprotein profile. 60 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Tonstad, 1995165 Boys and girls aged 6 164 To compare lipid levels and Family history: Parental history of cardiovascular to 16 with family history of premature disease based on hospital or physician report. heterozygous FH in cardiovascular disease in Parental lipid levels before treatment obtained Norway referred to lipid children with FHC210G, null from charts. History of early cardiovascular clinic or identified alleles, or no detected mutation, disease in 2nd degree relatives based on family through screening while taking into account parental report. families with known lipid levels, apoE polymorphism, Substance use: Those > age 12 asked about FH. FH diagnosis: LDL and anthropometric and dietary smoking in absence of parents. Alcohol use self- receptor mutation or 1 characteristics. reported (age 11+). prior TC >6.7 mmol/L Cardiovascular fitness: and if a parent had Anthropometric measures: Sum of skinfolds: tendon xanthomas triceps and subscapular sites. Percent body fat and/or TC > 7.8 calculated from sex-specific regression equation mmol/L with triglyceride with age and sum of skinfolds. Sexual maturation: levels <3.0 mmol/L Tanner scale. prior to treatment with Diet: Dietary instruction following National lipid-lowering drugs. Cholesterol Education Program guidelines. No specific weight-loss plans. Unexpected 24-hour diet recall at clinic visit, 4-day diet record. Venous blood: Samples following 8-10 hour fast. TC, TG, HDL. LDL calculated by Friedwald formula. 61 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Tonstad, 1995165 X NE NE Boys: LDL correlated with 24-hour recall cholesterol intake (p=.04) TG correlated with 24-hour recall sucrose intake (p=.004) 62 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Tonstad, 1995165 Multiple regression analysis: LDL related to FH parent's Multiple regression analysis: cholesterol levels, percent body fat, pubertal stage, and non- LDL related to percent body fat. FH parent's cholesterol level (p<.01), together explaining 40% Boys TG related to percent body fat of variance in LDL (CI 25%-55%). and percent dietary energy from HDL related to FH parent's HDL and to pubertal stage, sucrose explaining 17% (95%CI, 6% - 31%) of variance in HDL. Girls TG related to percent body fat, explaining 7% of variance (95%CI, 0.2% - 27%) 63 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Tonstad, 1995165 NE NE 64 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Twisk, 2001166 Subjects were from the 181 To investigate the development Biological: Blood - 10 ml for cholesterol and HDL- longitudinal Amsterdam of biological CHD risk factor C testing; Other - blood pressure; Skinfold Growth and Health clustering and to analyze the thickness was measured using four skinfolds: Study, who ranged in influence of lifestyle parameters triceps, biceps, subscapular, and suprailiac; age from 12-15 years on this clustering cardiopulmonary fitness - maximal oxygen uptake at the initial screening relative to body weight and a mean age of 27 Lifestyle: dietary intake - questionnaire of fat years at the final follow- intake, carbohydrate intake, Keys-score, and up alcohol intake; smoking - questionnaire of amount of tobacco smoked per week; physical activity - time spent on physical activities in relation to school and work, organized & unorganized sports, other leisure time activities, etc were measured during a structured interview. 65 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Twisk, 2001166 X Physical activity was inversely NE Heavy alcohol consumption was inversely related to the clustering score related to the clustering score (RR = 0.79, (RR = 0.89, p<0.01) p=0.01) but only for males. There was a weak inverse relationship for fat intake (RR = 0.95, p=0.05) and a weak positive relationship for carbohydrate intake (RR 1.05, p=0.05) 66 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Twisk, 2001166 NE Skinfolds and oxygen uptake were significantly related to TC:HDL (p<0.01, p<0.05, respectively). 67 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Twisk, 2001166 NE NE 68 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Twisk, 1998167 Males and females 181 To analyze longitudinal Six repeated measurements over 15 years, initially age 13. relationships between BMI/SSF starting at age 13. Participants healthy and biological and lifestyle risk Activity: Structured interview assessing total time students in Amsterdam factors for CHD. in activities, with a total weighted activity score secondary school. covering the 3 months prior. Substance use: alcohol use measured by self- report with diet. Smoking behavior questions in separate questionnaire. Cardiovascular fitness: maximal oxygen uptake measured on a maximal test using a treadmill with speed of 8 km/h and increasing slope. Anthropometric measures: BMI, sum of 4 skinfolds (triceps, biceps, subscapular, suprailiac). Lean body mass [(100-%body fat)xbody weight] (body fat estimated from SSF) Diet: cross-check dietary history interview Venous blood samples: TC, HDL 69 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Twisk, 1998167 XNENENE 70 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Twisk, 1998167 NE TC male: related to SSF (B=0.12, CI 0.07-0.18, p<0.01) TC female: related to SSF (B=0.07, CI 0.03-0.11, p<0.01) TC: related to BMI (B=0.05, CI 0.02-0.08, p<0.01). No relationship with lean body mass. HDL and SSF; HDL and BMI; HDL and lean body mass: no significant relationship TC:HDL ratio: related to SSF (B=0.08, CI 0.05-0.12, p<0.01); to BMI (B=0.04,CI 0.01-0.07, p<0.01); no relationship with lean body mass. 71 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Twisk, 1998167 NE NE 72 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Twisk, 1996168 Subjects were from the 181 To investigate the relationship Biological: Blood - 10 ml for cholesterol , HDL-C, longitudinal Amsterdam between the longitudinal and ratio of TC to HDL-C testing; Skinfold Growth and Health development of TC, HDL, and thickness was measured using four skinfolds: Study, who ranged in TC:HDL ratio and the evolution of triceps, biceps, subscapular, and suprailiac; age from 12-15 years lifestyle measures . cardiopulmonary fitness - maximal oxygen uptake at the initial screening relative to body weight and a mean age of 27 Lifestyle: dietary intake - fat intake, ratio of years at the final follow- polyunsaturated fatty acids and saturated fatty up acids, carbohydrate intake, cholesterol intake, and alcohol intake; smoking - questionnaire of amount of tobacco smoked per week; physical activity - time spent on physical activities in relation to school and work, organized & unorganized sports, other leisure time activities, etc were measured during a structured interview. vanLenthe, 1998169 Initial recruitment from 182 To investigate the association Anthropometric measures: Trunk skinfold 1st and 2nd grade of a between the change in a central measurements (subscapular, suprailiac) and secondary school in pattern of body fat and blood extremity measures (biceps, triceps) used to Amsterdam, the pressure and lipoprotein levels in create a subscapular/triceps skinfold ratio (S/T Netherlands who were healthy males and females age ratio) as an indicator of central pattern of body fat. invited to participate in 13 - 27. Venous blood samples: Non-fasting TC, HDL. a study on growth, Calculated TC/HDL. development and risk indicators for cardiovascular disease. 73 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Twisk, 1996168 X Daily physical activity was NE TC is inversely influenced by the ratio of positively related to HDL levels intake of polyunsaturated fatty acids to (p=0.38) saturated fatty acids (p=0.001) but is positively correlated with the intake of cholesterol (p=0.002). As well, the key scores are positively correlated with TC levels (p<0.001). TC:HDL ratio was positively correlated with the intake of carbohydrates (p=0.022). Alcohol consumption was positively related to HDL levels (p=0.001) and negatively related to TC:HDL ratio (p=0.005). No other dietary parameters were found to be significant for TC, HDL, or TC:HDL ratio. vanLenthe, 1998169 XNENENE 74 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Twisk, 1996168 NE NE vanLenthe, 1998169 NE Males: >S/T associated with < HDL (Beta -0.14, CI -0.24 to - 0.05, p<0.05). No association between change in S/T and change in TC/HDL Females: No significant association between change in S/T and lipoprotein levels (TC, HDL). 75 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Twisk, 1996168 NE Smoking: inversely related to HDL (p=0.007) and positively related to TC:HDL ratio (p=0.003) vanLenthe, 1998169 NE NE 76 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used van Stiphout, 1985170 Residents of a 458 To identify the distribution and Activity: Self-report via questionnaire by children suburban Dutch determinants of serum total over 10 community aged 5 to cholesterol and HDL in children Oral contraceptives: Females age 10+ were 19 who were a subset from a general Dutch population. asked about of those invited to Anthropometric measures: Quetelet index (ratio participate in a study of of body weight over height squared). Females risk factor for age 10+ asked about menstruation cardiovascular disease Diet: Self-report via questionnaire. Coffee and (EPOZ). alcohol used assessed Venous blood samples: non-fasting, TC, HDL of children and parental TC Smoking: Children aged 10+ asked about in questionnaire Ward, 1980171 74 volunteer families 74 To identify familial and Child measures taken at age 2.5 from a previous study environmental variables related Anthropometric measures: Triceps skinfolds on cord blood to cholesterol levels in children at Diet: Mother provided written diet record for child cholesterol. 74 children age 2.5. for 3 weekdays. Dietary history and food evaluated at age 2.5 frequency record obtained for each parent. years. Blood samples: 5-ml cholesterol, hemocrit 77 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet van Stiphout, 1985170 X TC: No relationship to physical NE TC: No relationship to coffee activity level consumption. Ward, 1980171 X NE NE Stepwise multiple regression: TC: Previous breast-feeding (vs formula), current dietary poly-unsaturated/saturated fatty acid ratio, maternal alcohol and fat consumption, paternal cholesteryl esters, and maternal hemocrit (r=0.53). Free cholesterol: similar predictor variable set (r=0.52) Esterified cholesterol of children: predicted by history of breast-feeding, child's current protein and caloric intake, maternal total and esterified cholesterol, and proteinuria during pregnancy (r=0.69). 78 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures van Stiphout, 1985170 TC: For both fathers and mothers positively related to TC of Body weight and Quetelet index: + relationship with TC for children (p<0.001 except for males in quartile 2, where p< males and females > age 15 (p< 0.05). No relationship for 0.005). those < 15 years of age. Body weight: Negative relationship with HDL in males age 15+ (p<0.05 for quartile 3, p<0.01 for quartile 4). Quetelet index: Similar negative relationship as body weight. (No p value given) Ward, 1980171 In stepwise multiple regression NE TC: related to paternal cholesteryl esters and maternal hemocrit (r=0.53) Free cholesterol: similar predictor variable set (r=0.52) Esterified cholesterol of children: predictors included maternal total and esterifies cholesterol (r=0.69) 79 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous van Stiphout, 1985170 Age: Decrease in total cholesterol between ages 9 and 16 NE Menarche: No relationship with TC Ward, 1980171 NE NE 80 Evidence Table 2. Risk Factors Population and Study, year Setting N Aims Measures Used Wong, 1992172 Pediatric-office-based 1081 To evaluate the importance of Total cholesterol, height, weight, blood pressure. cholesterol screening family history factors and Family medical history questionnaire, dietary program in 2-20 year television viewing habits of habits, exercise, TV viewing. old children; 87% children for white, 4% Hispanic, 2% hypercholesterolemia. Asian, 7% other NA = Not applicable, NE = Not evaluated, RR = Risk ratio, X = Indicates that an adjusted analysis is present (sensitivity analysis). 81 Evidence Table 2. Risk Factors Results Adjusted Aerobic Capacity / Cardiovascular Study, year Analysis Activity Fitness Diet Wong, 1992172 X Exercise and activities were not NE Families who used only extra-lean or no associated with cholesterol hamburger (vs lean or regular level status. hamburger) were more likely to have Watching 2-4 or >4 hrs of children with a cholesterol level of 200 television daily to have mg/dL or higher (RR 2.5, p<.01). Other cholesterol level of 200 mg/dL dietary habits (type of milk, skin or no skin or higher (p<.01). on chicken, etc) were not associated with cholesterol level status. 82 Evidence Table 2. Risk Factors Results continued Study, year Family History Overweight or other Biological Composition Measures Wong, 1992172 Family history of a high cholesterol level but not premature BMI not associated with cholesterol level status. myocardial infarction was associated with an increased likelihood of high cholesterol (RR 1.62, p<.05). 83 Evidence Table 2. Risk Factors Results continued Other Other Study, year Biological Miscellaneous Wong, 1992172 Age and blood pressure level were not associated with NE cholesterol level status. 84 Evidence Table 3. Screening Strategies Family History Determination Method & Study, year Venue N Laboratory Method Definition Bachman, Routine screening via 1160 Random serum cholesterol test at Any 1st or 2nd degree relative (parent, sister, 1993175 family history children Kaiser Permanente Regional brother, uncle, aunt, grandparent) with heart attack, questionnaire at pediatric ages 2-18 laboratory using methods approved angina or stroke Bell, 1990115 Screening event via 1140 fifth Non-fasting capillary blood samples Survey of parents for: history of heart attach, schools in Scotsdale AZ grade (Autoclix) collected; total cholesterol stroke, angina, high cholesterol, smoking and high students measured by wet-chemistry methods blood pressure in child's parents and grandparents. using three Abbott Vision analyzers. Age at first occurrence for heart attack, stroke, angina Bistritzer, Hospital-based screening 107 children Fasting venous samples, IL Monarch Family history determined from father's histories of 1995176 of children of father (ages 2-20) analyser MI or CHD (coronary artery stenosis > 70% of 1+ hospitalized for of 40 men major branches) before age of 40. Fathers were myocardial infarction < (age 31-41) patients particular Israeli hospital. age 40 and women Israel 1 Evidence Table 3. Screening Strategies Study, year Screening Results Cost Comments Bachman, Of 1160 who completed questionnaires, 529 (46%) had positive family history of NR 1993175 CHD prior to age 55 or high cholesterol; 369 (70%) of these had a random cholesterol level drawn as recommended. 93 subjects had TC>185mg/dL; 58 of these (62%) completed follow-up testing for lipid panel. Of the 58 who had a panel drawn, 25 (43%) had LDL>126mg/dL and were offered a 3 or 6 week nutrition program. Of the 25, 9 (35%) enrolled in the program. Bell, 1990115 1201 returend parental permission and family history forms. 34 children were NR absent or refused testing. 27 samples were hemolyzed or inadequate. Thus, 1140 students had cholesterol samples measured. 25% had TC>185 mg/dL (4.75 mmol/L) and 13% had levels greater than 200 mg/dL (5.2 mmol/L). Family history of parent or grandparent with high cholesterol, hypertension or vascular disease had the highest sensitivity (77%) but lower specificity (24%). Parent or grandparent high cholesterol had sensitivity of 53% and specificity of 62%. The group with negative familiy history included 40% of all the levels greater than 185 mg/dL and 36% of all the levels greater than 200 mg/dL. Bistritzer, 3 men died from MI and 3 refused to participate, so 34 men remained. NR 1995176 8 (24%) had normal values. 26 (77%) had 1+ abnormal values including: TC > 6.24 mmol/l, TG > 2.55 mmol/l, LDL-C > 4.42 mmol/l, HDL-C < 0.91 mmol/l. 6 were obese (BMI > 30 kg/m2) 15 mothers (38%) had significant hyperlipidemia (TC > 6.24 mmol/l, TG > 2.55 mmol/l, LDL-C > 4.42 mmol/l, HDL-C < 0.91 mmol/l, or combination). 6 of the mothers were obese. Children' s mean values: TC 4.68 mmol/l, TG 1.40 mmol/, LDL-C 3.00 mmol/l, HDL- C 1.18 mmol/l. Altogether 42% of the children had significant hyperlipidemia (above 95%ile). 1 child was obese (BMI > 27 kg/m2). 2 Evidence Table 3. Screening Strategies Family History Determination Method & Study, year Venue N Laboratory Method Definition Boulton, Screened samples of 200 2 year 2 year olds had non-fasting capillary Family history obtained from parents of children with 1979131 children various ages olds; 385 4 samples; 4 year olds had fasting TC and/or LDL >95th percentile for age; early CHD taken from a cohort study, year olds; capillary samples; school children defined as <65 yrs, hypercholesterolemia in a parent a preschool health 230 students had fasting venous samples was defined as a TC>251.5 mg/dL program, and a school- ages 8-17 based screening program Davidson, School-based screening, 420 of 786 Capillary testing, Kodak CT-60 Sibling, parent or grandparent history of MI, bypass 1989177 4th graders invited analyzer; repeated with serum test surgery, stroke, diabetes, high blood pressure, children (Reflotron analyzer) for those with smoking and high cholesterol participated TC>200 mg/dL Dennison Otsego County, NY 10,457 N/A Questionnaire asking household adults >age 20 to 1994133 population survey children completed information about cardiovascular ages 2-19 disease, early cardiovascular disease ( 3 Evidence Table 3. Screening Strategies Study, year Screening Results Cost Comments Boulton, Case finding rate was greatest for the older age group: For 2 year olds, 15 were NR Population and method 1979131 >95th percentile with 12 having family history of early CHD; For preschoolers, 20 differences make had TCD>95th percentile, 9 with family history of early CHD; For primary school, 4 comparisons of these three children had TC>95th percentile, 2 of whom had family history of CHD; For groups (ages 2, 4 and 8-17) secondary school, 7 students had TC>95th percentile of whom 5 had family history problematic) information and all had a family history of early CHD. Davidson, 33% of children the upper decile of cholesterol levels had a family history of early < Cost <$2 per test. 1989177 age 56 MI in grandparents, none had family history of early CHD in parents. 83% of those asked to return were confirmed to have TC>185mg/dL Dennison Most of the family history of early coronary heart disease was not due to coronary NR other than, 1994133 heart disease in the parents (only 5.8% had parents with early coronary heart "parents who disease). belonged to an Children living in two-parent households were more likely to report a known family HMO were much history of early coronary heart disease than children living in single-parent more likely to households (41.8% vs 25.8%, p<0.001) and twice as likely to have known parental have had their hypercholesterolemia (18.8% vs 9.5%, p<0.001). cholesterol levels Authors adjusted their estimate of children qualifying for cholesterol screening in measured than order to account for the younger age (<55 yrs) for definition of early coronary heart parents with disease used by the NCEP. With this adjustment, 29% of children from 2-parent private group families (reduced from 41.8%), and 18.4% of children from one-parent families insurance, (reduced from 25.8%) would qualify for lipoprotein analysis. Overall 38% of this Medicaid, or no population would qualify for cholesterol screening. health insurance. 35% of children had incomplete or unavailable family health history and/or unknown parental cholesterol status. 4 Evidence Table 3. Screening Strategies Family History Determination Method & Study, year Venue N Laboratory Method Definition Dennison, Bogalusa 2127 Lipid Research Clinic protocol Questionnaire completed by parent/guardian, asking 1989126 children whether child's natural mother or father had a ages 3-17 history of "high blood pressure," heart attack," "stroke," or "sugar diabetes." Four years later a questionnaire asked parent's age at first heart attach for 108 of the 144 parents who had previously reported a history of heart attack. Diller, 1995117 Cincinnati Myocardial 241 families A singleTC and LDL-C determined Screening indicators were: history in parent or Infarction and Hormone and 501 in a CDC and NHLBI standardized grandparent of documented myocaridal infarction, (CIMIH) Family Study children laboratory. angina, pectoris, peripheral vascular disease, ages 2-19 sudden death, angiographically demonstrated coronary artery disease, ballon angioplasty, or coronary bypass grafting before age 56. Discretionary indicators were: HTN, obesity, diabetes, smoking, steroid use, high fat diet. Faigel, Mass screening of college 9,938 Students required to submit AAP criteria (heart disease, CAD or stroke prior to 1992178 students students cholesterol results upon age 50) ages 17-21 matriculation; those >200 mg/dL (95th percentile approximately) were referred to student health for lipid profile 5 Evidence Table 3. Screening Strategies Study, year Screening Results Cost Comments Dennison, Only 40% of white children and 21% of black children with elevated LDL levels NR 1989126 (>95th percentile) had a parental history of vascular disease. Diller, 1995117 Prevalence of screening indicators was 47.8%. NR Prevalence of discretionary indicators was 54.7%. 9.9% had high LDL-C (>130mg/dL)and 15.1% had borderline LDL-C levels (110- 130mg/dL). Of the 58 children with elevated (borderline or high) LDL-C, 7 had no identifiable screening indicators, 38 had at least one of the discretionary indicators. Diet alone would have identified 29 children with elevated LDL-C values. Parental cholestoler concentration >240 mg/dL would have identified 31 children with elevated LDL-C. Faigel, 427 of the 873 students with cholesterol levels >200 mg/dL completed fasting lipid Cost per case 1992178 profile; 344 of these had abnormal levels (TC>200, TG>180; LDL>160 or HDL<30). identified was Others had testing done by their PCP (data not included). 43 of the 344 students $212. with abnormal lipid profiles met AAP criteria with family history of early CHD. 6 Evidence Table 3. Screening Strategies Family History Determination Method & Study, year Venue N Laboratory Method Definition Gagliano, All new patients at an 224 of 248 Total cholesterol obtained at time of Questionnaire administered to patient by physician; 1993118 adolescent/young adult new patients visit; those with TC>185mg/dL (85th positive if any member of the family (1st and 2nd clinic associated with a participated percentile) advised to have repeat degree relatives) had early myocardial infarction children's hospital fasting lipid panel (Kodak Ektachem (<50 for men, <60 for women) or elevated lipids; 700) negative if no history or unknown. Parent interviewed if present at clinic visit. Garcia, Routine surveillance of 6500 ages 3-Non-fasting venipuncture with Family history of premature MI or known 1989179 pediatric patients age 3- 18 (mean Reflotron analyzer, confirmed with hypercholesterolemia 18 6.4 years) fasting lipid profile if >200 mg/dL (95th percentile) Goff, 1991186 Routine screening at well- Ages 4-19 Non-fasting capillary bloodsample NR child visits in the office of with Reflotron analyzer 4 pediatricians. Griffin, Routine screening of 1,005 Non-fasting serum cholesterol: Parent or grandparent with hypercholesterolemia, 1989119 children in pediatric office children commercial laboratories or physicin sudden death, gout, peripheral vascular disease, practices. ages 2-13 offices (desk-top analyzers). any family history of obesity. Any family history of Children with TC>175mg/dL (75th coronary artery disease, various combinations. percentile) were asked to return for fasting lipid panel. 7 Evidence Table 3. Screening Strategies Study, year Screening Results Cost Comments Gagliano, Family history based on adolescent interview had 36.4% sensitivity and 69.1% NR 1993118 specificity for hypercholesterolemia (TC>185mg/dL); family history based on parent interview was 65% sensitive and 45.7% specific; combined history was 45% sensitive and 69% specific. Of the adolescents with elevated cholesterol levels, 16 (48%) would have been missed if screening was performed only based on family history (including 1 of 2 patients with cholesterol >240 mg/dL). Of the 33 patients with TC>185 mg/dL, only 13 (39%) returned for fasting lipid profile. Of these 13 patients, 6 had LDL>125 mg/dL, HDL<35, or TC/HDL>4.5. Garcia, 552 (8.5%) had cholesterol > 200mg/dL; Of those 487 completed the fasting lipid NR 1989179 profile and 375 had LDL>130 (95th percentile); Of these 42 had LDL>190mg/dL; 90 had LDL between 160-189 mg/dL and 243 were between 130-159mg/dL Goff, 1991186 In 12 month period, 138 or 791 boys (18.5%) and 100 of 165 girls (16.3%) had TC NR screening values that exceeded the 90th percentile for gender. In ther first 6 months, 238 patients were identified. 37% of boys and 38% of girls received follow-up by a second cholesterol check, a lipoprotein analysis or both. Overall, 37.4% of those with sex-sepcific TC > 90th percentile and 43% of those with sex-specific TC >95th percentile had f/u within 6 months Griffin, Using a family hisotry of early coronary arter disease or hhpercholesterolemia NR 1989119 (parent or granparent) would have detected 51% of children with elevated LDL-C. 8 Evidence Table 3. Screening Strategies Family History Determination Method & Study, year Venue N Laboratory Method Definition Hammond, Routine screening of 593 Non-fasting venipuncture NR 1994180 children in seven primary 295 (50%) schools in ages 5-6; Cantebury England 298 (50%) selected from a range of ages 8-9 socioeconomic backgrounds Heyden, Cholesterol education and 1910 tested Reflotron instruments (Boehringer- NR 1991181 screening in 5 North in 1987 Mannheim Diagnostics, Indianapolis, Carolina high schools, 1845 tested Indiana), using a ages 15-17 in 1988 finger stick blood sample USA Lannon Routine screening of all 439, ages 2- Capillary testing, Kodack DT-60 NR 1992130 children at well-child visits 15 analyzer, if high advised to return for in six-member pediatric fasting lipid panel (LRC approved practice lab) 9 Evidence Table 3. Screening Strategies Study, year Screening Results Cost Comments Hammond, Of 593 consent forms, 443 parents (74.7%) gave positive parental consent. Of the NR Paper looks at feasibility of 1994180 443, 13 children (2.9%) were absent for the venipuncture, 21 children (4.7%) testing; no data on levels. refused, for 17 (3.8%) there was a technical failure, 2 (.05%) felt unwell or Note significant difference in unspecified. Venus sample obtained from 390 (88%). 375 (84.6%) were usable for failure rate of phlebotomists. analysis. Lower parental response rates were from schools with lower socioeconomic backgrounds. Heyden, In both males and females, the 2 schools that included teacher involvement had NR Paper examines methods of 1991181 significant decreases in cholesterol levels over a one year period. Males in first cholesterol education. school (n= 219) levels decreased from 145.1 to 129.7 (p<0.05). Males in second Interventions: teacher school (n=261) levels decreased from 151.1 to 131.6 (p<0.05). Females in first involvement plus testing vs school (n=280) levels decreased from 162.7 to 143.3 (p<0.05). Females in second no teacher involvement plus school (n=321) levels decreased from 165.2 to 151.7 (p<0.05). There were no testing. significant differences in the 3 other schools that did not include teacher Teacher involvement involvement. described as: discussing cholesterol, nutrition, heart disease; demonstrated how to read package labels; counseled on cafeteria food choices, follow up discussion after testing, etc. Testing occurs first in 1987, then again to evaluate methods in 1988. Lannon Of 439 screened, 134 (31%) had TC>175 or LDL>110 mg/dL (AAP criteria). Of Cost of lipid profile 1992130 those with abnormal screening tests, 63 (47%) returned for lipid panel; of those $13 returning, 54 (86%) were confirmed to have hypercholesterolemia. Of those 134 children, 23 (17%) had negative family history and thus would have not been screened according to AAP/NCEP criteria. 10 Evidence Table 3. Screening Strategies Family History Determination Method & Study, year Venue N Laboratory Method Definition Lansing, Mass screening at state 69 people Capillary testing, Reflotron analyzer. Not part of screening program 1990182 fair ages 2-19 Manchester, RCT of mass screening of 735 of a Non-fasting blood sample, Kodak Any relative with CAD or stroke 1989183 college students vs. total 1,117 Ektachem analyzer; confirmed for selective screening (all students those with TC>180 mg/dL (75th sophomores with family ages 17-19. percentile) history of heart disease) Muhonen, Mass screening of high 599 students fasting venous blood sample, LRC Students completed family history questionnaire at 1994120 school students grades 9- from a total processing methods home, with help of parents. Family history assessed 12 in Muscatine, IA of 1,024 for parent or grandparent with fatal or nonfatal MI, students, angina, bypass surgery, HTN, high cholesterol or ages 14-20 obesity. Classifications: no CHD, early CHD (between ages 30-55 for at least one parent or grandparent); later CHD (after age 55 in at least one parent or grandparent); and unsure (incomplete forms, missing information or unsure about at least one parent or grandparent. 11 Evidence Table 3. Screening Strategies Study, year Screening Results Cost Comments Lansing, Well-accepted. 3 found to have moderate risk, 3 to have high risk (not defined) Cost per person 1990182 found with elevated cholesterol was $40 (includes adult cases); $140 per person who visited his/her physician. Manchester, 735 (62%) of students had cholesterol tested; no difference in family history Cost per case 1989183 between those who had cholesterol measured and those who did not, although the confirmed was aim was to get those with family history. 60 students had confirmatory $121 for mass measurement; 50 remained over 180mg/dL. Only 10% of those with a risk factor screening vs $624 chose to participate. for selective screening. Muhonen, Of 1,024 targeted students, 599 had both family history questionnaire and lipid NR 1994120 measurements completed. 19% indicated a parental history of hyperlipidemia; 26% had unsure responses. 2.5% had a parental history of early CHD. 12 Evidence Table 3. Screening Strategies Family History Determination Method & Study, year Venue N Laboratory Method Definition Polonsky, Pediatric lipid clinic at 182 mean Fasting levels of TC, HDL-C and TG NR 1993184 Children's Hospital age 9 (18 were drawn at the clinic and Medical Center, months - 20 analyzed at the one central Cincinnati, OH years) laboratory. Resnicow, School-based screening 1166 Non-fasting capillary samples Questionnaire sent home to parents asking them 1993122 in 4 elementary schools, analyzed with Kodak Ektachem DT about their cholesterol levels (asked to provide grades 1-6 60. specific value if known) Rifai, 199662 African American 260, ages fasting venous samples, Ektachem Questionnaire completed by patients adolescents at the 12-20 years 700 analyzer, CDC certified lab. Adolescent medicine and (74% Ambulatory Clinic: 260 of female) 201 enrolled Shea, 199059 Columbia Study of 108 fasting venous blood samples Posi5tive if child's mother had HTN, diabetes, Childhool Activity and Hispanic analyzed for TC and HDL-C and TG obesity, hyperlipidemia or family hsitory of Nutrition (recruited mainly children who in a Columbia University research premature CHD or hyperlipidemia. No data on risk from General pediatrics had lipids laboratory calibrated to Centers for factors obtained from the fathers. Also positive if Group Practic at measured Disease Control standards. history of hyperlipidemia, premature CHD, MI or Presbyterian Hospital, sudden death in child's parent or aunt, uncle, or New York) granparents 13 Evidence Table 3. Screening Strategies Study, year Screening Results Cost Comments Polonsky, 16 (9%) subjects had normal lipid levels and classifications. Of those (59 subjects, NR 1993184 32%) with elevated LDL-C levels, 16 had concurrent elevated TG or depressed HDL-C. In the TG:HDL-C group (54 subjects) 46 had TG > 95th percentile. For subjects with HDL-C below the 5th percentile, 5 had borderline TG (75th to 95th percentile) and 3 had normal TG levels. Resnicow, Using a parental cut point of 200 mg/dL, 34% of children would have been NR 1993122 screened. 40.3% and 48.5% of children with TC values >170 mg/dL and >200 mg/dL would have been identified. Using NCEP/AAP recommended parental cut- point of 240 mg/dL, 9% of children would have been screened and 10.3% of children with TC values >170 mg/dL and 200 mg/dL would have been identified. If clinicians screen only children with one parent who reports a total cholesterol>240mg/dL, between 90-93% of children with elevated total cholesterol (>170) would be missed. Rifai, 199662 37% of population would have met NCEP/AAP criteria for screening. Of 95 patient NR meeting screening criteria 25 (26%) had LDL>110 mg/dL (75th percentile); 11(12%) had LDL>130mg/dL (90th percentile). 66% of those with LDL>110mg/dL and 60% of those with LDL>130mg/dL would have been missed under current screening guidelines, even including those with unknown family history. Shea, 199059 Using AAP definition of family history, sensitiviy was 57% and specificity was 59% . NR When family history defined according to NIH consensus conference and the AHA, senstiivity was 46% and specificity was 70%. 12 of 106 would have been misclassified on the basis of TC level (LDL-C would have placed them in another risk category). 14 Evidence Table 3. Screening Strategies Family History Determination Method & Study, year Venue N Laboratory Method Definition Skovby, Copenhagen school 2085, ages Non-fasting capillary blood sample Parents completed a questionnaire on incidence of 1991185 system 6-8 obtained between 8am-12pm. chest pains or coronary occlusions in themselves or their parents, siblings, aunts, uncles including age. Positive family history defined as angina pectoris or myocardial infarction Steiner, Kaiser Permanente 1001, 12-21 Random serum cholesterol test Questionnaires were completed about risk factors 1991124 Teenage Health Centers years old performed at one central laboratory. and family history. Used AAP criteria from 1985 in Los Angeles, CA and 1988. Sveger, School-based 4000 quest- Blood samples collected after Family history questionnaire that asked parents 200079 questionnaire, with lipid ionnaires; overnight fast from high risk child whether they or the child's grandparents has levels in those who were 2199 and parent(s). Centrifugation serum suffered premature CAD (before age 50 in father or high risk and their (55%) were was separated and frozen at -20o C grandfathers; or before age 55 in the mother or parents, ages 10-11 returned for later analysis. grandmothers). Definition of genetically linked dyslipidemias: 1) LDL-C > 189.48 mg/dL + parent has hypercholesterolemia or died of premature CAD. Hypercholesterolemia in parent = above, or if TC > 290.03 mg/dL + myocardial infarction occurred before age 50 (male) or 55 (female) in first degree relative 2) child has increase of LDL-C + apolipoprotein B, or only apolipoprotein B, and a parent with hypercholesterolemia and/or hypertriglyceridaemia. 15 Evidence Table 3. Screening Strategies Study, year Screening Results Cost Comments Skovby, 3025 questionnaires distributed, 2657 returned, 2166 consented to screening with NR 1991185 capillary blood measurement. Of the 2,657 questionnaires completed, 398 families (15%) disclosed a history of early cardiovascular disease. Steiner, Only 30 of 64 (sensitivity = 47%, specificity = 74%) subjects with NR 1991124 hypercholesterolemia were identified by the 1985 AAP criteria, while 40 of the 64 (sensitivity = 62%, specificity = 60%) subjects were identified by the 1988 AAP criteria. Combining the two criterias (1985 & 1988 AAP criteria) 44 of the 64 (sensitivity = 69%, specificity = 53%) subjects were identified. Sveger, Family history of premature CAD in parents or grandparents was identified in 208 NR 200079 families. 182 agreed to participate. Blood samples obtained from 175. 89 children had normal tests (LDL-C < 131.48 mg/dL, HDL-C > 34.80 mg/dL, TG < 124 mg/dL, apolipoprotein B < 1.00 g/l, and Lp(a) < 300 mg/l. Of the remaining: 48 had isolated increase in Lp(a). 23 had non-hereditary abnormalities of low LDL or high HDL cholesterol or apolipoprotein B. 15 were suspected to have genetically determined dyslipidaemias or a combination of risk factors: possible familial hypercholesterolemia in 4; possible familial combined hyperlipidemia in 5; hereditary low HDL in 3; combination of high LDL and Lp(a) in 3. Also, possible familial hypercholesterolemia was detected in 8. 16 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Amundsen, 200241 Diet Randomized, double-To assess effect of SE- Study period: Subjects recruited from blind crossover enriched spread on serum 25 weeks patient register at Lipid Oslo, Norway lipids, lipoproteins, with run-ins Clinic of National carotenoids, fat-soluble and Hospital in Oslo, vitamins, and physiologic washouts Norway variables in children with Two 8 week FH aged 7 - 12 years. treatment periods. 1 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Amundsen, 200241 Not specifically noted. Implied: 38 Mean age: 10.5 + 1.7 1. Parent with hypercholesterolemia 2 non-study-related drop- 50% female 2. Diagnosed with "definite" or "possible" outs; 1 drop out because Cholesterol (mmol/L) heterozygous FH the amount of spread total: 7.01 + 1.26 3. Healthy, with no clinical symptoms of required to be consumed LDL: 5.39 + 1.42 hypercholesterolemia was too large. HDL: 1.37 + 0.32 Triacyglycerol (mmol/L) 0.56 + 0.22 2 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Amundsen, 200241 Exposure: Measures: Main Clinical Outcomes: In a double-blind crossover, with two 8- Plasma total, LDL and 1. Plasma LDL decreased by 10.2% (p=0.003) with week interventions, 38 children with FH HDL cholesterol treatment vs control. consumed 18.2 + 1.5 g SE spread/day, Triacylglycerol 2. Total cholesterol and apolipoprotein B concentrations corresponding to 1.60 + 0.13 g SE apolipoprotein (apo) A-1 decreased by 7.4% (p=0.007 and p=0.020, respectively) spread, or a control spread. Blood and apo B with treatment vs control. samples were analyzed at the start and Serum concentrations of 3. No changes with treatment in HDL, triaclglycerol or end of each diet period. Subjects carotenoids, retinol, and a-apolipoprotein A-I. consumed a recommended American tocopherol 4. Serum concentration of lipid-adjusted lycopene Heart Association Step I diet during Blood analyses decreased by 8.1% (p=0.015) with treatment, with no both intervention periods. other carotenoid changes. 5. Lipid-adjusted retinol and a-tocopherol concentrations increased by 15.65% (p<0.001) and 7.1% (p=0.027), respectively. 6. Alaninetransaminase increased with treatment, which was explained by a significantly lower starting concentration in the treatment vs control period. 3 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Amundsen, 200241 ALT increased in SE diet 16.8% (p = 0.04). No subjects had Small N Fair plasma concentration of ALT outside the reference Many children used supplements or concentration at any time. Starting concentration of ALT was vitamins during the study. significantly lower in the SE diet period compared to control. These variables were not analyzed Serum levels of lycopene and B-carotene decreased with SE. separately. After adjustment for lipid changes, differences in B-carotene FH was confirmed by mutation analysis in disappeared; lypocene concentration was still 8.1% lower only 25/38 subjects. (p=0.015) in SE compared to control. Some included children may not have had Serum levels of retinol were higher in SE, and after lipid FH, as baseline total and LDL cholesterol standardization: 15.6% (p<0.001) higher than control. concentrations were significantly higher in After lipid standardization, ) a topopherol was 7.1% higher in SE the confirmed FH group. than in control (p=0.027). Authors note results may be influenced by different starting concentrations in SE spread than control. 4 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Becque, 1988221 Diet and RCT of exercise, To determine the incidence 20 weeks Recruited via exercise diet and behavior of coronary heart disease advertisements in local combination change. risk factors and the effects newsletters and Michigan of a 20 week diet and newspapers. exercise intervention program on risk factors including serum lipids. 5 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Becque, 1988221 Inclusion criteria: 36 obese adolescents Girls (n-21): 1. Body weight and triceps skinfold greater randomly assigned to mean age 12.8+0.3 year. than the 75th percentile according to three groups: diet plus Height 156.3+1.5 cm NHANES data behavior (n=11), weight 71.3+3.6 kg exercise plus diet and behavior change (n=11) boys (n=15) or control (n=14). mean age 12.7+0.05 year. Height 157.4+2.3 weight 70.8+4.7 kg. 6 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Becque, 1988221 Control group was encouraged not to TG, HDL, TC Results in mg/dL and reported as mean+SEM post change their basic lifestyle systolic and diastolic BP treatment vs post-control. When NS stated means even Diet plus behavior change: subjects body weight, % fat and within group changes were not significant. met with dietician and behavior maximum to uptake therapist once/week. Moderate dietary Control group (n=14): restriction based on ADA exchange TG increased 117.8+15.8 to 122.2+13.6, NS program with caloric intake set to elicit HDL increased 29.5+1.6 to 32.0+1.5, NS the loss of 0.45-0.90 kg/week (1-2 TC decreased 1776.9+10.7 to 167.1+9.0, NS lbs/week). Behavioral treatment included record Exercise, diet and behavior change group (n=11) keeping stimulus control, changing TG: decreased 135.8+16.5 to 91.6+18.4, NS topography of eating and reinforcing HDL increased 35.3+2.3 to 43.4+2.2 (p<0.05 for within altered behavior. group difference, P<0.05 for between group difference in post-intervention levels). Exercise plus diet and behavior change: TC decreased 170.6+10.2 to 149.3+9.2, NS diet and behavior were same as above, exercise program was 50 minutes 3 Diet and behavior change group (n=11) days/week. Each session included TG decreased 117.3+13.9 to 99.9 +17.4, NS warm-up and aerobic activity, HDL increased 34.5+2.7 to 38.4+3.0, NS increasing from 15-40 minutes per TC decreased 181.2+12.3 to 171.9+11.4, NS session of the course of the fist 7-8 weeks. HR maintained at 60-80% of age-predicted maximum. 7 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Becque, 1988221 NR Fair 8 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Boreham, 2000230 Exercise RCT of stair To evaluate the effects of a 7 weeks Previously healthy climbing. 7 week progressive stair- female student Belfast, Ireland climbing program on volunteers ages 18-22, cardiorespiratory fitness sedentary but previously and lipid levels in young healthy. women Clauss, 200538 Drug RCT NR 24 weeks Post-menarchal girls with heFH 9 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Boreham, 2000230 Responded that they exercised or played 25 people enrolled; data 22 females, ages 18-22 sport on one or no occasions each week. reported on 22 women Students at the University of Belfast. who completed the study. Clauss, 200538 HeFH; baseline LDL required was 160-400 54 girls ages 10-17 at 54 girls ages 10-17 mg/dL and parental history of familial least 1 years post- hypercholesterolemia. menarche 10 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Boreham, 2000230 Stair-climbing group underwent a 7 1. Serum lipids Pre-post results in mmol/L, mean + SEM after 7 weeks week program, progressing from one 2. Cardiorespiratory ascent/day in week 1 to 6 ascents/day fitness (VO2 maximum Stair climbers: in weeks 6-7 on a public access and HR) TC decreased 4.93+0.25 to 4.53+0.24, p<0.05 within staircase Monday-Friday. Staircase 3. Anthropometry group was 199 steps with a total vertical HDL increased 1.24+0.09 to 1.48+0.14, p<0.05 but displacement of 32.8 meters. Subjects p<0.01 for interaction between groups and p<0.05 for were instructed to distributed their pre-intervention differences between this group and the efforts evenly over the working day and control. to log completed ascents (ie: could be done in separate short bursts). Completed ascents were 75% of the target over the 7 weeks period. Control group was asked to maintain their normal lifestyle over the experimental period. Clauss, 200538 Lovastatin 20 mg for 1st 4 weeks, then Serum lipids, apo-B Mean % change from baseline at week 24 in ITT 40 mg thereafter population: TC: -22.4 for lovastatin vs +3.6 for placebo LDL: -29.2 for lovastatin vs +2.5 for placebo HDL +2.4 for lovastatin vs +4.8 for placebo TG: -22.7 for lovastatin vs -3.0 for placebo Apo B: -24.4 for lovastatin vs +6.4 for placebo 11 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Boreham, 2000230 None reported 3 subjects (2 controls and 1 stair climber) Poor withdrew because of illness or academic pressures. Clauss, 200538 "Adverse reactions to lovastatin were generally mild and Unpublished data from lovastatin package Fair transient." Patients treated with lovastatin had an adverse insert and confirmed through contact with experience profile generally similar to that of patients treated Merck, Inc. with placebo. In this limited controlled study, there was no detectable effect on menstrual cycle length in girls. Adolescent females should be counseled on appropriate contraceptive methods while on lovastatin therapy." 12 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Couture, 199866 Drug RCT To determine whether the 6 weeks FH patients at university Laval University nature of the LDL receptor research clinic Lipid Research mutation effects the Clinic, Quebec City, response to simvastatin. Canada To describe the different responses to simvastatin of plasma lipids, lipoproteins, and apoprotein levels among 3 genetically differentiated groups of heterozygous FH children and adolescents. 13 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Couture, 199866 1. Heterozygous FH patients with 1of 3 63 37 boys and 26 girls mutations in the LDL receptor gene. Aged 8-17, mean 12.55 2. Aged 8 - 17 3. Weight > 27 kg 4. Plasma levels persistently above 95th percentile for age and sex while maintaining a lipid-lowering diet. 5. No concomitant conditions, such as diabetes, anorexia, thyroid disorders 14 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Couture, 199866 Individual screening with medical Total Cholesterol Compared to placebo, drug reduced total cholesterol, history, physical exam, interview with HDL cholesterol LDL, and total apoB levels at all time measurements dietician, and blood sampling 6 weeks LDL cholesterol (p<0.0001). HDL levels increased in all groups with drug prior to study entry. All lipid-lowering apoA-I versus placebo (p=0.003 to 0.30), and triglyceride levels medications discontinued at screening apoB decreased (p=0.009 to 0.10). Multiple regression entry. Four weeks placebo run-in, apoE genotypes analyses suggested that 42% of the variation of LDL followed by randomization to double- triglycerides response to drug is due to variation in the mutant LDL blind active treatment, 20 mg/d BMI receptor locus, apoE genotype, and BMI; 35% of simvastatin or placebo for 6 weeks at variation in HDL response was due to gender and 3:1 ratio. Compliance verified by tablet baseline HDL level. count at weeks 0, 2, 4, and 6. Patients questioned about adverse or unusual signs or symptoms at all clinic follow-up visits (weeks 0, 2, 4, and 6). Patients counseled by dietician to follow American Heart Association phase I diet throughout trial. 15 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Couture, 199866 None reported Small sample size Fair 16 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Davidson, 199667 Diet Double-blind, To investigate the 6 weeks of Children aged 6 - 18 randomized cross- hypocholesterolemic effect treatment in with in the Chicago area over study of psyllium-enriched cereal each with in children with condition hypercholesterolemia documented (8 weeks diet recruited by family hyperlipidemia. stabilization, screening and public 6 weeks appeal. treatment 1, 6 weeks wash-out, 6 weeks treatment 2) de Jongh, 200269 Drug Double-blind, To evaluate LDL- 48 weeks of 173 boys and girls with placebo controlled cholesterol-lowering intervention heFH multi-center RCT efficacy, overall safety, after 4 week 9 international sites tolerability and the run-in influence of growth and pubertal development of simvastatin in a large cohort of boys and girls with heterozygous familial hypercholesterolemia (heFH). 17 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Davidson, 199667 1. Males and premenarcheal females aged 6 - 25-32 randomized ; 2 25 children, ages 6-18. 18. dropped due to inability Gender and mean age not 2. LDL > 90th percentile for age and sex at to consume amount of reported, although females baseline. cereal required, 1 were premenarchal. 3. Height and weight < 75th percentile discontinued for 4. No chronic medical condition, such as noncompliance, 2 diabetes. discontinued for family 5. Triacylglycerol < 300 mg/dL problems. Of the 27 who 6. Any lipid-lowering agents, including dietary completed study fiber supplements, discontinued > 6 weeks protocol, 2 excluded from before 1st qualifying blood draw analysis because baseline LDL varied markedly de Jongh, 200269 1. Age 10 - 17 173 98 boys and 75 girls 2. LDL-C 158.55 - 398.3 mg/dL 3. Parent with confirmed diagnosis of heFH. 4. Boys: Tanner stage II or above Girls: postmenarchal for at least 1 year before study initiation Exclusion: homozygous familial hypercholesterolemia, secondary hyperlipidemia 18 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Davidson, 199667 8 week NCEP diet stabilization followed Fasting lipid profile Psyllium cereal vs control reduced total (12.1 mg/dL, by blood sampling and random (baseline, ever 6 weeks, p=0.03) and LDL cholesterol (10.9 mg/dL, p=0.01). LDL- assignment to active or control cereal week 19) cholesterol decreased 7% in active vs control groups. for 6 weeks. 6 week wash-out, then 6 Compliance assessment Response to treatment differed between subjects week crossover treatment period. Dietary assessment consuming psyllium during period 1 vs period 2. A period- Active: 58 g of psyllium-enriched cereal (including computerized by-time interaction was found for total cholesterol daily. Each serving had 3.2 g soluble nutrient analysis at weeks (p=0.02). Subjects consuming psyllium had an average fiber. Control: matching placebo cereal -8, -1, 6, 12, and 19) decrease in total cholesterol of 19.1 mg/dL during the 1st without psyllium. Subjects instructed to period and 5.6 mg/dL during the 2nd period; LDL consume two 28-g boxes of cereal daily decreased 17.1 mg/dL during the 1st period and 5.2 for a total daily dose of 6.4 soluble fiber mg/dL during the 2nd period. from psyllium. Dietary compliance assessed by cereal box count every 3 weeks, questioning of children individually and with parents, and by 3 day dietary records. Subjects on NCEP diet throughout study. de Jongh, 200269 After 4 week placebo/diet run-in, Total cholesterol, LDL-C reduced at all time points by drug vs placebo randomization to active treatment or triglycerides, LDL-C, HDL- (p<0.001). At 24 weeks, LDL-C reduced 38.4% to 125.29 matching placebo in 3:2 ratio and C, ALT, AST, CK, apoB, mg/dL in drug vs 1.2% reduction in placebo (p<0.001). stratified by sex. Simvastin (drug) apoA-1, adrenal Total-C, VLDL-C, and apoB also reduced vs placebo at started at 10mg/day and increased at 8 hormones, gonadal all time points (p<0.001). TGs reduced at weeks 8, 16 week intervals to 20 and then 40 hormones, pituitary and 48. HDL-C and apoA-1 increased for all weeks, and mg/day for the remainder of the study hormones and FSH. In were only significant relative to placebo at week 24 (period 1, 24 week duration) and for a girls, B -human chronic (p<0.05). No changes in either treatment group noted in 24 week extension (period 2). Office gonadotropin was either treatment group at week 24 or 48. visits every 4 weeks. Menstrual cycle measured each visit. monitored throughout study period and Tanner staging used to assess pubertal development. 19 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Davidson, 199667 None reported. "Only one child complained of gastrointestinal 8/25 subjects who completed the study Fair effects with slight abdominal bloating" were < 80% compliant during 1 or both treatment periods. Response to treatment differed between subjects consuming psyllium between period 1 and 2. Small sample size. de Jongh, 200269 No serious adverse effects. In period 1, 1 discontinuation, due 223 assessed for eligibility, 175 randomized Good to mononucleosis. In period 2, 4.7% of drug vs 3.4% of placebo - 69 to placebo, 106 to control. 5 of placebo reported > 1 drug-related clinical AE. (Difference not discontinued: 1 lost to follow-up, 2 withdrew significant). Total drug group clinical adverse effects included: consent, 2 for other reasons. 5 of drug abdominal pain (3), chest pain (1), flatulence (1), mylgia (2), discontinued: 1 to mononucleosis, 1 to headache (4), sleep disorder (1), weight gain (1), pruitus (1). protocol deviation, and 3 to consent Total drug group lab adverse effects: increased ALT (3), withdrawal. 101/106 drug completed period increased AST (3), increased CK (1). 1 and 83/106 completed period 2 (78%). No deleterious effects on growth or pubertal development. 56/69 placebo(81%) completed period 2. 20 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting de Jongh, 200268 Drug RCT with an To determine whether 28 weeks; HeFH patients ages 9- additional arm of simvastatins improves dosage 18 non-FH controls. endothelial function in doubled USA. children with familial every 8 hypercholesterolemia (FH) weeks de Jongh, 200340 Diet Double blind To determine whether 4 weeks 41 prepubertal HeFH randomized cross- prepubertal HeFH children each period, children between ages 5- over trial are characterized by 2-6 week run- 12 and 20 non-FH impaired endothelial in, 6 week controls. function and whether short- wash-out The Netherlands. term intervention with plant between sterols can improve treatment endothelial dysfunction. periods 21 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics de Jongh, 200268 Children with heterozygous FH as defined by: 50 heterozygous FH FH simvastatin group Mean 1. LDL>95 percentile for age/gender; 2. children plus 19 non- age: 14.6+2 Female gender: Documented family history of hyperlipidemia affected controls 13 (42%) Race: NR. FH with LDL>95 percentile for age/gender after placebo group Mean age: treatment; 3. Personal diagnosis by 14.6+2.5. Female gender: 11 detection of mutation at the LDL receptor (50%) Race: NR. gene de Jongh, 200340 Children with HeFH selected as follows: 1) 41 children with HeFH. mean age 8.2 years LDL>95 percentile for age and gender; 2) 20 males, 21 females documented family history of hyperlipidemia with LDL>95th percentile for age and gender prior to treatment or 3) personal diagnosis of FH by detection of mutation in the LDL receptor gene. Excluded if : girls post-menarchal, boys with Tanner stage later than stage 1, smoking, current use of vasoactive medications or dietary supplements, any other serious illness (HTN, DM). 22 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results de Jongh, 200268 50 heterozygous FH children 1. FMD (flow mediated Mean absolute change in FMD was higher in the FH randomized to simvastatin or placebo in dilitation) of brachial simvastatin group vs FH placebo group: 3.9%+4.3% vs. 3:2 ratio. An additional 19 healthy non- artery at baseline and 28 1.2%+3.9%, p=0.05. Mean absolute reduction in TC FH siblings also used as controls. weeks. 2. TC, TG, LDL, mmol/l (FH simva -2.16+1.04 vs -FH placebo 0.05+1.17, HDL; 3. AST, ALT, CK at p=0.0001); increase in HDL mmol/l (FH simvastatin each visit for safety; 4. 0.05+0.17 vs FH placebo -0.05+0.22, P=0.080); Height, weight and BP. decreased in LDL (FH simva -2.13+0.99 vs FH placebo - 0.05+1.06 mmol/l, p=0.0001); and decrease in TG (FH simvastatin -0.19+0.37 vs FH placebo -0.10+0.54 mmol/l, p=0.041). de Jongh, 200340 All HeFH children also were on a low- Endothelial function by Mean+sd in mmol/L for HeFH children randomized to saturated -fat, low-cholesterol diet (step flow-mediated filation placebo vs plant sterol spread (baseline data are I). (FMD) at R brachial combined for both P and S groups): artery. TC decreased from 7.30+1.51 to 7.06 for placebo vs 6.27+1.12 for sterol group, p<0.001. LDL decreased from 5.68+1.51to 5.4+1.137 for P vs 4.58+1.13 for sterol group, p<0.001 HDL increased from 1.25+1.25 to 1.29+0.29 in P vs 1.31+0.31 in S, p=0.594. TG increased from 0.74(range 0.46-2.20) to 0.90+0.40 vs. 0.85+0.36, p=0.476. FMD was significantly lower in the FH placebo group than in the healthy controls (7.2%+3.4% vs 10.1%+4.2%, p<0.005) with no difference in baseline vessel size. Among the HeFH children, FMD in the P and sterol treated groups were not significantly different (7.2%+3.4% P vs 7.7%+4.1%, p=0.592). Again, baseline vessel size between P and S were similar. 23 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating de Jongh, 200268 "In the present cohort, no toxicity or serious adverse or side 28 in FH simvastatin group, 22 in FH Poor effects wre reported by the children during the course of this placebo group, 19 non-FH controls. study. However, the duration of the present study is too short to draw conclusions with regard to the saftey of long-term use of simvastatin in children." de Jongh, 200340 "FH children did not report any adverse effects throughout the Compliance was 97% in both P and S Good study." groups based on returned empty tubs of margarine. 24 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Dennison, 1993223 Drug randomized cross- To assess whether 4-5 weeks children ages 5-17 with over trial of psyllium psyllium was effect at each period LDL>110 mg/dL already lowering TC and LDL with a 2 week being treated with diet, values in children beyond wash-out recruited from Pediatric the reduction achieved by period Lipid Control Center and diet alone between from private pediatricians in the vicinity. 25 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Dennison, 1993223 Children with LDL>110 mg/dL after treatment 25 children began the 11 males; 9 females with diet. study; 4 did not return for age 11.1+3.8 years the final blood drawing (3 because of parental difficulties - timing, logistics; 1 boy because he refused to eat the cereal and have his blood redrawn); 1 second blood sample was lost. 26 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Dennison, 1993223 Diet consisted of <30% calories as total 1. Dietary intake (food Results for psyllium vs control cereal (also given are fat, <10% as saturated fat and <200 mg diary kept by subjects and results for each group, change from baseline with the dietary cholesterol/day. Total calories their parents) control and psyllium cereals separately). Values are in enough to ensure adequate growth. 2. Serum lipid levels mmol/L, mean +SEM. Subjects instructed to eat two 28gm 3. Serum levels of TC decreased by 0.01+0.10, NS servings (1 0z or 2/3 c each) of the vitamin D, folic acid, HDL increased 0.03+0.04, NS control cereal (5 mg water-insoluble vitamin A, vitamin E, TG decreased 0.68+0.32, p<0.05 wheat fiber per serving) or the psyllium calcium, iron, TIBC and LDL increased 0.10+0.11, NS cereal (3 gm of water-insoluble fiber transferrin saturation and 3 game water-soluble fiber per 4. Height, weight, Serum iron was higher after the psyllium cereal than serving). Cereals were similar in subscapular and triceps after the control (92 vs 75 mg/dL, p=0.04). appearance and provided by same skin-fold thickness No changes in subscapular or triceps skin-fold manufacturer. thicknesses. Height and weight increased equally during consumption of control and psyllium cereals. None of vitamin or mineral levels tested were lower after the psyllium cereal. 27 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Dennison, 1993223 One child (a 17 year old girl reported a transient increase in baseline lipid levels not different significantly Poor loose stools (3 /day) with the control cereal with resolved with for those (n=5) who did not complete the discontinuation and reappeared with a second challenge of the study. control cereal and resolved post discontinuation. Compliance was 82% for both cereals. Most families were unaware of which cereal was in use (2/3 of children or parents incorrectly identified the psyllium containing cereal). 28 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting DISC Collaborative Diet RCT To assess the efficacy and 3 years Prepubertal boys and Research Group, 6 centers safety of lowering dietary girls recruited from 1995219 intake of total fat, saturated public and private fat, and cholesterol to elementary schools, by decrease low-density mass mailings to lipoprotein cholesterol members of an HMO, (LDL-C) levels in children and from pediatric practices 29 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics DISC Collaborative 1. Girls aged 7 years, 10 months - 10 years 1 663 (362 boys, 301 girls) Mean age: boys 9.7 Research Group, month; boys aged 8 years, 7 months - 10 girls 9.0 1995219 years, 10 months 2. Average of 2 screening LDL-C values > 80th and < 90th percentile for age and sex. 3. No medication or medical condition that could effect growth or blood cholesterol 4. No behavior problems in child or family likely to reduce adherence 5. Prepubertal 6. No plans to move within the 3 study years 30 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results DISC Collaborative Randomized to diet (334) or usual care Height and serum ferritin At 3 years, dietary total fat, saturated fat, and cholesterol Research Group, (329). levels at 3 years levels decreased significantly in the diet group compared 1995219 Diet: Behavioral intervention to promote LDL-C, total serum to control (p<.001 for all). LDL-C decreased in diet and adherence to a diet providing 28% of cholesterol levels, control groups by 15.4 mg/dL and 11.9 mg/dL, energy from total fat, less than 8% from triglycerides, HDL-C, respectively. With adjustment for baseline level and sex saturated fat, up to 9% from dietary assessment, and imputing values for missing data, mean differences polyunsaturated fat, and less than skinfold thickness, body between the groups was -3.23 mg/dL (95% CI -5.6 - 0.5 75/mg (1000 kcal) per day of circumferences, and mg/dL) (p=.02). cholesterol (<150 mg/day). Family blood pressure at 1 and 3 oriented, based on social learning and years. No differences between groups in adjusted mean height, social action theory. 1st visit: eating Red blood cell folate, serum ferritin levels, or other safety outcomes. pattern assessed and personalized serum zinc, retinol, and program developed. In the 1st 6 albumin at 3 years months: 6 weekly and then 5 biweekly Annual Tanner staging, group sessions augmented by 2 height, weight and individual visits of children with their psychosocial family members. In the 2nd 6 months, 4 assessments group and 2 individual sessions. Years 2 and 3: group individual and maintenance sessions held 4-6 x per year with monthly phone contacts between sessions. Usual care: Public educational publications on heart-healthy eating provided. Parents informed if child blood cholesterol high - no specific recommendations to see physician given. 3-year lipid results provided, with referral as clinically warranted. 31 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating DISC Collaborative No differences between groups in adjusted mean height, serum Authors state randomization yielded Good* Research Group, ferritin levels, or other safety measures. comparable treatment groups, although 1995219 Serum ferritin decreased in both groups and the intervention baseline diet group had lower group had slightly lower mean ferritin concentrations than did polyunsaturated fat (p=.03), higher vitamin usual care at year 3 (p=0.08). In both groups, mean ferritin B6 (p=.04) zinc (p=.02), a higher proportion concentrations remained above the 75th percentile for age and with household income < $20,000 (p=.002), sex. and scored higher on the Child Depression Risk of consuming less than 2/3 of the RDA was significant for Inventory (p=.09) vitamin E at all visits (baseline OR: 1.009; year 1 OR: 1.007; Low drop-out rate year 3 OR: 1.007; p<0.0001 for all visits); for zinc at all visits for boys (baseline OR: 1.004, p<0.05; year 1OR: 1.003, p<0.02, and year 3 OR: 1.004, p<0.003) and girls (baseline OR:1.007, p<0.001: year 1 OR: 1.008, p< 0.0006 and year 3 OR: 1.005, p < 0.003). 32 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Engler, 2003224 Drug RCT To determine the effects of 6 months Not reported antioxidant vitamin therapy and the NCEP-II diet on endothelial functions well as on surrogate biomarkers for oxidative stress and inflammation. 33 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Engler, 2003224 1. Children aged 8 - 21 with hyperlipidemia. 15 (8 boys, 7 girls). 4/19 Not reported 2. FH (LDL >130 mg/DL and a parent 19 were excluded due to diagnosed) or FCH (LDL >130 mg/dL, or noncompliance with triglycerides >150 mg/dL or both, and at least vitamin therapy. 1 parent presenting with any of these phenotypes). 3. No systematic illness with or without secondary hyperlipidemia. 4. No current smoking. 34 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Engler, 2003224 After 6 weeks of adherence to NCEP-II Brachial artery baseline Antioxidant vitamin therapy and NCEP-II diet improved diet, patients randomized to 6 weeks of diameter, flow-mediated FMD of the brachial artery compared with baseline vitamin C (250 mg) and vitamin E (200 dilation, cholesterol (total, (p<0.001) without an effect on biomarkers for oxidative IU) twice daily or placebo. Following 6 LDL, VLDL, HDL), stress (antibodies to epitopes of oxidized LDL, F2- weeks of wash-out, patients crossed- triglycerides, lipoprotein, isoprostanes, 8-hydroxy-2'-deoxyguanosine), over to alternate treatment group for 6 OxLDL antibody titer, inflammation (C-reactive protein), or levels of weeks. All followed NCEP-II diet OxLDL-E06 levels, asymmetric dimethylarginine, an endogenous inhibitor of throughout the trial with the support of ADMA, F2-isoprostanes, nitric oxide. nutritional counseling. Dietary intake C-reactive protein, urinary The primary effect of the NCEP-II diet was an 8% assessed with 3 non-consecutive 24- 8-OH-2'dG, blood reduction in LDL (p<.001), with no other significant hour recalls at each visit. Nutrient intake pressure, BMI effects of diet or antioxidants on other lipid levels or assessed. Compliance with biomarkers of oxidative stress, inflammation or ADMA. supplements determined through pill counts. Blood and urine samples obtained at baseline and every 6 weeks. 35 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Engler, 2003224 None reported Very small sample size Poor No demographic/setting description 36 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Engler, 2004234 Drug Randomized cross- To determine whether diet 6 weeks Not reported over trial (NCEPII) plus placebo or each pase diet plus docosahexaenoic with 6 week acid (DHA) affects washout endothelial function in period children with FH or FCH. between 37 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Engler, 2004234 Children ages 8-21 with FH or FCH. FH 20 subjects ages 9-19, baseline BMI for entire group = characterized by LDL>130mg/dL and parent 12 with FH and 8 with 21+4; no race data reported diagnosed with the disorder. FCH was FCH. characterized by elevated levels of LDL (>130 mg/dL) or triglycerides (>150 mg/dL) or both, and a parent presenting with 1 of these 3 phenotypes. Excluded were those with chorinc systemic illness with or without secondary hyperlipidemia and current smoking. 38 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Engler, 2004234 After 6 weeks of adherence to NCEP-II Primary outcome was Cross-over results presented in aggregate. However, diet, pts were randomized to 6 weeks of brachial artery flow- the repeated measures anova indicated no main effect placebo or DHA (all continued on diet). mediated dilation. or order and no order by phase interaction "indicating Subsequently, following a 6 week Secondary outcomes that presenting DHA first or placebo first had no effect." washout period, the groups crossed were lipid levels FMD increased after DHA supplementation compared to over. placebo (p<0.012) TC was higher for DHA plus diet compared to diet plus placebo: 297+81 vs. 286+85 mg/dL TC increased for DHA plus diet compared to diet alone: 263+79 vs. 297+81mg/dL (pre-post comparison, not randomzied, p>0.006). LDL was higher for DHA plus diet compared to placebo plus diet: 229+85 vs. 216+98 mg/dL. HDL was lower for DHA plus diet compared to placebo plus diet: 51+13 vs. 52+11 mg/dL. TG were lower for DHA plus diet compared to placebo plus diet: 119+71 vs 131+94 mg/dL 39 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Engler, 2004234 Not Reported Poor 40 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Engler, 2005236 Drug Randomized cross- To determine whether diet 6 weeks Not reported over trial (NCEPII) plus placebo or each pase diet plus docosahexaenoic with 6 week acid (DHA) affects washout endothelial function in period children with FH or FCH. between 41 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Engler, 2005236 Children ages 8-21 with FH or FCH. FH 20 subjects ages 9-19, baseline BMI for entire group = characterized by LDL>130mg/dL and parent 12 with FH and 8 with 21+4; no race data reported diagnosed with the disorder. FCH was FCH. characterized by elevated levels of LDL (>130 mg/dL) or triglycerides (>150 mg/dL) or both, and a parent presenting with 1 of these 3 phenotypes. Excluded were those with chorinc systemic illness with or without secondary hyperlipidemia and current smoking. 42 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Engler, 2005236 After 6 weeks of adherence to NCEP-II Lipoprotein subclass Results report an shift toward larger, more buoyant LDL diet, pts were randomized to 6 weeks of profile and HDL particles without significant changes in placebo or DHA (all continued on diet). quantitative total cholesterol amount. Subsequently, following a 6 week Cross-over results presented in aggregate. washout period, the groups crossed LDL was higher for HAD+diet compared to placebo + over. diet (226+86 vs 216+92). LDL-1 and LDL-3 had signficant differences vs. placebo. HDL-2 was significantly higher for DHA+diet vs placebo +diet as compared to diet alone. No significant change in HDL overall, or in triglycerides. 43 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Engler, 2005236 Not Reported Poor 44 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Epstein, 1989217 Diet and RCT To assess the effects of 6 months Community exercise weight change over 6 combination months on serum lipid changes 45 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Epstein, 1989217 1. Children ages 8-12 56 children randomly mean age 10.5 in treatment 2. One child and one natural parent who assigned to diet (n=18), groups, 10.3 in control were >20% of ideal weight for age, sex and diet plus life-style % overweight: 42 in treatment height. exercise (n=19) or no group, 50 in control group 3. Child's triceps skinfolds >95th percentile treatment control (waiting for age and sex list) (n=19). 4. No history of psychiatric contact for the children 46 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Epstein, 1989217 Diets were set between 3800- Weight, height, fitness, Results for the two intervention groups (diet or 5000kJ/day and monitored by and serum lipids diet+exercise instruction) were combined for total anutrtionis to maintain nutrient treatment group n=35 and control n=16. 51/56 adequacy. completed the trial. 9 did not have 6 month serum data Information on diet, exercise, stimulus because they refused venipuncture. Values given as control, reinforcement, modeling and mean (SD) in mmol/L after 6 months: contingency contracting was presented Change in total cholesterol was -0.27 (0.47) for to parents and their children in eight intervention vs +0.09(0.50) for control, p=0.03 weekly sessions followed by four Change in HDL was 0.20 (0.16) for intervention vs +0.06 monthly sessions. (0.13) for control, p=0.007 Change in triglycerides was -0.55 (0.53) for intervention vs -0.12 (0.30) for control, p=0.01 Change in weight -3.6 kg (-17.4% change in percent overweight) for intervention vs +5.2 kg (-0.8% change in % overweight). 47 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Epstein, 1989217 NR other than 9 children refusing venipuncture. Poor 48 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Ferguson, 1999225 Exercise Randomized cross- To examine the effect of 8 weeks Obese children living in over trial of 4 exercise training on obese community near the months of exercise children Medical College of training (ET) Georgia; recruited via community fliers and community/hospital newspapers 49 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Ferguson, 1999225 Inclusion criteria: 81 children; 2 were Mean age, s.d= 9.5+1.0 yrs children need to have a triceps skinfold>85th dropped for medical mean height= 141.9+8.7 cm percentile for gender, ethnicity and age and reasons prior to mean weight= 57.6+17.7 kg not be involved in any other weight control or randomization. 79 black=44 subjects; white=37 exercise program, and not be restricted as to children randomized. subjects; Asian=1 subject physical activity; all females were menarchal 26 boys and 53 girls 50 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Ferguson, 1999225 2 days diet recall TC, HDL, LDL, TG, LDL- Results at months 4 (prior to cross-over) in mmol/L, least Exercise program offered 5 days/week; C size, Lp (a), ApoA-1, square mean+s.e. children were paid $1/session for each APoB, insulin, glucose, TC: ET group (n=40) decreased from 4.29+0.08 to day of attendance and given prizes for GHb(%), %fat and 4.04+0.08; control group (n=37) decreased from satisfactory participation (maintaining a submaximal HR. 4.63+0.08 to 4.27+0.08. HR>150 bpm). Transportation was HDL: ET group (n=40) increased from 1.14+0.03 to provided to the program after school. 1.19+0.03 vs control group did not change (n=37) Each session was 40 minutes long with 1.35+0.03 to 1.35+0.03 20 minutes of aerobic activity and 20 LDL: ET group (n=39) decreased 2.48+0.08 to minutes of games that were designed to 2.33+0.08 vs control group decreased 2.79+0.08 to assure continuous activity. 2.45+0.08 TG: ET group (n=39)decreased 1.15+0.01 to 0.95+0.01 vs control group (n=36) increased 0.98+0.01 to 1.10+0.01 Lp (a): ET group (n=37)0.15+0.01 to 0.13+0.01 vs control group (n=37) decreased 0.16+0.01 to 0.12+0.01g/L Apo A-1: ET group (n=39) increased 1.10+0.03 to 1.24+0.04 vs control (n=37) decreased 1.31+0.03 to 1.28+0.03 ApoB: ET group (n=39) decreased 0.77+0.01 to 0.76+0.02 vs control (n=37) decreased 0.79+0.01 to 0.74+0.02. Results significant for the change in TG ("p-0.02 for the group x time interaction over entire 8 months). Results also given for both groups at the end of the cross over 51 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Ferguson, 1999225 NR 76 children tested at month 4; 70 children Fair tested at month 8. 3 subjects withdrew from the group that had no exercise in the first 4 months; 3 children withdrew from each group during the 2nd 4 month period. 52 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Gold, 1991231 Diet RCT To evaluate serum 4 weeks School-age children in apolipoprotein and lipid the U.S. profile effects of an oat bran supplemented low-fat diet in children with elevated serum cholesterol. 53 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Gold, 1991231 1. Cholesterol level over 185 mg/dL 49 Mean age: 10 years 54 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Gold, 1991231 Subjects given American Heart ApoA1, apoB, total OBS had decreased apoB (-9mg/dL) vs control Association Step 1 Guidelines (fat cholesterol, LDL, HDL, (2mg/dL), p=0.05. No significant differences between intake limited to 30% of total calories and triglycerides groups in total cholesterol, apoA, LDL, HDL, or serum and cholesterol to 300 mg/day), then triglycerides. randomized to control or oat bran supplemented (OBS) groups. OBS consumed an estimated 38 grams of oat bran/day in the form of cereals and snack bars for 4 weeks. 55 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Gold, 1991231 No adverse side effects or symptoms reported. No information on compliance, withdrawals, Poor intent-to-treat. Very brief write up. 56 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Gylling, 199539 Diet Randomized cross- To assess the effect of 6 weeks 14 children with HeFH over trial of diet sitosterol rapeseed each arm Helsinki, Finland margarine vs. rapeseed margarine without sitosterol 57 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Gylling, 199539 Diagnosis of FH established in children and 14 children with HeFH 7 boys, 7 girls in one parent mostly by DNA technique and 1 child with HoFH (2 Mean age 9.1+1.1 yr (2-15) yr old) 58 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Gylling, 199539 Children were randomized to replace 24 TC, VLDL, IDL, LDL, Results combined after cross over, p-value assumed grams of their normal daily fat intake by HDL, HDL/LDL ratio, given notation of "significant change between groups." the same amount of a rapeseed oil-rich HDL2, HDL3, mean+SE in mmol/L, pre values were combined. margarine with or without sitostanol phospholipids and ester (M vs MS) . All children had subfractions, triglycerides TC: decreased from 7.68+0.36 to 7.62+0.32 in M group previously been advised to use a low and subfractions. vs 6.81+0.34 in MS group (10.6% lower in MS group, animal fat-low cholesterol diet rich in p<0.05). monoenic fatty acids for years. Protocol results in daily consumption of LDL decreased from 5.64+0.35 to 5.47+0.30 in M group 3 grams of free sitostanol in MS group. vs 4.65+0.32 in MS group (15% lower in MS group, p<0.05). VLDL increased from 0.21+0.05 to 0.26+0.06 in M group vs 0.25+0.07 in MS group (3.8% lower in MS group, NS). HDL increased from 1.17+0.07 to 1.20+0.07 in M group vs 1.25+0.08 in MS group (4.2% higher in MS group, NS). TG increased from 0.87+0.10 to 1.03+0.13 in M group, vs 0.92+0.12 in MS group, (10.7% lower in MS group, NS). Data reported separately for HoFH boy 59 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Gylling, 199539 Children noted no difference in taste between the two Poor margarines. Report "good compliance," and "well tolerated." 60 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Kang, 2002226 Exercise RCT of exercise. To test whether high 8 months Obese 13-16 year old Atlanta, GA intensity physical training youths were recruited. would have a favorable effect on components of the insulin resistance syndrome in obese adolescents 61 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Kang, 2002226 Inclusion criteria: 80 obese youths enrolled White boys: n=10, mean age 1) triceps skin-fold greater than the 85th in two cohorts and 14.5years. percentile for gender, ethnicity and age randomized within white girls: n=15, mean age 2) no involvement in any other weight control gender and sex to one of 15.3yr. or exercise program three groups: black boys: n=16, mean age 3) no restriction with respect to ability to lifestyle education alone 14.1yr. engage in physical activity. (LSE); LSE+moderate black girls: n=39, mean age intensity physical training 15.2 yr. (PT); and LSE + high intensity PT. 62 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Kang, 2002226 Physical training was offered 5 1. TC, HDL, LDL, Lp (a), Pre-post changes in mmol/L least-square mean+SEM: days/week for 8 moths except during apo A1, ApoB, glucose, TC: LSE alone increased 0.24+0.10 vs LSE+PT the weeks that they were assigned to insulin, triacylglycerol decreased 0.03+0.09, p=0.066. lifestyle education (LSE). 2. Percentage of body fat HDL: LSE alone decreased 0.03+0.04 vs Lse+PT Physical training program involved an 3. Systolic and diastolic decreased 0.03+0.04, p=0.958. individual exercise prescription based BP VLDL: LSE alone increased 0.12+0.04 vs LSE+PT on baseline treadmill test, designed for decreased 0.04+0.04, p<0.001 1045 kJ/session (250kcal) regardless of TG: LSE increased 0.12+0.08 vs LSE+PT decreased whether it was high/moderate intensity. 0.22+0.08, p=0.002 The intensities corresponded to a mean Lp (a): LSE increased 2.45+4.79 mg/dL vs LSE+PT prescribed HR of 55-60% peak for increased 9.88+4.32, p=0.263 moderate intensity group and 75-80% Apo A1: LSE increased 10.51+4.31 vs LSE+PT of peak for high intensity group. increased 12.40+3.90, p=0.748 ApoB: LSE decreased 1.03+3.29 vs LSE+PT decreased 0.30+2.97, p=0.868. 63 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Kang, 2002226 Not Reported Poor 64 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Knipscheer, 199671 Drug double-blind, To assess the safety, 12 weeks, Lipid clinic randomized placebo tolerability and efficacy of following an The Netherlands controlled study, pravastatin for HeFH 8 week diet children stratified by and placebo age. 4 arms (3 run-in period. doses of pravastatin, 1 placebo) 65 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Knipscheer, 199671 Children with "known heterozygous FH" were 72 children with HeFH 25 male; 47 female eligible. HeFH defined as plasma LDL above 66/72 White; 5/72 Black. 95th percentile for age and sex during lipid- Mean age 11.9-12.1 across 4 lowering diet AND hypercholesterolemia in groups. siblings, parents or grandparents, or clinical manifestations of premature atherosclerosis prior to age 50 in 1st/2nd degree relatives. Exclusions: 1. Major surgery within the past 3 months 2. Use of medications interfering with lipid metabolism (anticonvulsants, oral contraceptives, corticosteroids, fibric acid derivatives or immunosuppressants). 3. Hepatic or renal dysfunction 66 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Knipscheer, 199671 Pravastatin 5, 10 and 20 mg/day 1. TC, LDL, HDL, TG, 1. Compliance was 93% Apo B100 2. At week 12: 2. TSH, cortisol, ACTH TC and LDL were significantly reduced in all pravastatin before and after treatment treated groups compared with baseline (p<0.05). LDL: 5 period mg/day - 20% decrease; 10 mg/day - 20% decrease; 20 3. Physical exam at entry mg/day - 30% decrease (from baseline LDL levels of and final visit. 247.49 mg/dL for P, 239.75 mg/dL for 5mg/day, 235.89 ALT, AST, T bilirubin, CK, mg/dL for 10 mg/day and 259.01 mg/dL for 20 mg/day. Alk phos. TC: 5 mg/day - 18% decrease; 10 mg/day - 18% decrease; 20 mg/day - 25% decrease (from baseline TC of 301.63 mg/dL for P, 297.76 mg/dL for 5 mg/d, 235.89 mg/dL for 10 mg/d, and 317.09 mg/dL for 20 mg/day. (all numbers estimated from graph). Plasma LDL cholesterol levels were not reduced below the 95th percentile for sex and age, except for 2 children in the 20mg pravastatin group, and 1 child each in the 5 and 10 mg pravastatin groups. Plasma HDL showed a mean increase of 11% (from baseline of 46.4 mg/dL) in the 20 mg pravastatin group (p<0.001), but was unchanged in the other groups. TG did not change as compared from baseline in any group. Plasma VLDL and Apo B100 levels were reduced in all pravastatin treatment groups (p<0.05 and p<0.001 respectively). 67 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Knipscheer, 199671 Total number adverse events P- 9, 5 mg/day - 3, 10 mg/day - 6, Good 20 mg/day - 1. List: rash, fatigue (P only), nose bleeding, headache, diarrhea (P only), dyspepsia (P only), nausea/vomiting, abdominal pain, myalgia (P only). Laboratory safety measurements did not show significant changes in any of the groups between the end of the treatment period and baseline. However CK abnormal in 8 of P , 6 of 5mg/day, 11 of 10 mg/day and 8 of 20 mg/day. Cortisol abnormal in 2 of P, 2 of 5 mg/d, 5 of 10 mg/d, 3 of 20 mg/d. 68 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Kuehl, 1993227 Diet RCT of parental and To identify and Children with TC >185 family education characterize a family- ages 2-15 (mean 7) USA based group intervention to lower plasma cholesterol in children. 69 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Kuehl, 1993227 Children referred by physicians who 295 children randomized SSI group: participated in a seminar on the risk factors to either a single session mean age 6.6 for atherosclerosis in children, identification intervention (SSI) or a gender: NR and management of high cholesterol in multi-session race: NR children. 14 pediatric practices referred 295 intervention (MSI) children. MSI group: mean age 7.6 gender NR race NR 70 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Kuehl, 1993227 1. SSI=one 90 minute nutrition 1. TC, TG, LDL, HDL Both groups had improvement in TC from baseline to the education session for patient, siblings 2. Caloric intake second visit: and parents (slides presentation, low fat 3. Diet composition TC: MSI group TC (n=90) decreased from 207.9+3.2 food prep and tasting, distribution of 4. Nutritional intake of to193+3.0 (p<0.0001; SSI group (n=87) TC decreased fruit and cereal). "Returned for calcium and iron from 206.1+3.3 to 198.5+3.3 (p<0.01). subsequent food sampling with 5. Growth HDL: MSI group decreased 53.7+1.4 to 51.6+1.4 nutritionist present" but no further (p<0.08). SSI group NR. formal education. LDL: MSI group decreased 137.0+3.6 to 124.9+3.3 2. MSI= four 90 minute sessions with (p<0.0001); SSI group NR. focus on food preparation for breakfast, TG: MSI group decreased from 88.5+6.8 to 83.5+5.1 (p snack, lunch and dinner. Received NS); SSI group NR. notebooks with nutrition information and recipes, incentives for attendance and Both groups had a significant decreased in the completion of behavioral contracts proportion of total calories obtained from fat. (eating low fat meal) Both groups had a significant increase in the proportion of total calories obtained from protein. Total calorie intake decreased in the MSI group but not in the SSI group: MSI 1464+34 to 1348+31 (p<0.002); SSI group 1358+33 to 1306+32 (p=NS). 71 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Kuehl, 1993227 Both groups maintained iron intake at over 87% and calcium Large and unequal drop out rates (35% for Poor intake at over 81% of the RDA throughout the study. % RDA of SSI and 16% for MSI) and analysis only use iron reported baseline to visit 2: MSI 103+5 to 98+5 (pNS); SSI those who completed the second blood 87+ 4 to 90+4 (p=NS). draw (215 of original 295); ie not intention to treat. All patients "grew linearly with a mean height increase of 2.0+0.3 inches in the MSI group and 2.0+0.1 in the SSI group. No assessment of intervention fidelity. Both groups had a mean weight increase of 4.1 lbs. Growth parameters not significantly different between the two groups. 18% of MSI and 20% of SSI had plasma TC values <170 and were excluded from the analyses because they didn't meet the initial referral criteria of TC > 185. 72 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Lambert, 199672 Drug RCT of drug To determine the efficacy, 8 weeks of Boys aged 17 or multicenter: 6 sites safety and tolerance of randomizatio younger with in Canada short-term administration n to meds vs. heterozygous FH of lovastatin in a male placebo; 4 pediatric population with wk run-in for severe FH, and to evaluate both groups. dose-response relationship. 73 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Lambert, 199672 1. Weight 27 kg or more 69 randomized, 10 mg/day group 2. Plasma LDL above 95th percentile for age 17 to 10 mg/day group Mean age: 12.5 (2.4) while on lipid-lowering diet 18 to 20 mg/day group Female gender: 0 3. History of unsuccessful treatment with bile 19 to 30 mg/day group Race: 94.1% white acid-binding resins 15 to 40 mg/day group 20 mg/day group 4. Family history of atherosclerosis at or Mean age: 12.7 (1.6) before age 50 Female gender: 0 5. Documented family history of Race: 100% white hyperlipidemia with LDL > 95th percentile for 30 mg/day group age and sex before treatment or a personal Mean age: 13.3 (2.7) diagnosis of FH substantiated by LDL Female gender: 0 receptor activity or mutation. Race: 94.7% white 6. No significant concomitant conditions 40 mg/day group 7. Weight and height not 3 - 97th percentile Mean age: 12.9 (2.7) for age Female gender: 0 Race: 93.3% white 74 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Lambert, 199672 After pre-study visit and lab tests, all Total cholesterol All doses reduced total cholesterol, LDC, and ApoB (P < lipid-lowering medications were LDL .0001 for all). Mean percent change from placebo to end discontinued at least 8 weeks before Apolipoprotein B of treatment varied from -17% to -29% for total study start, with no lipid-lowering Safety cholesterol values, from -21% to -36% for LDL, and from therapy allowed other than that Compliance -19% to -28% for ApoB. Dose response relationship, permitted by study protocol. Placebo with improved results up to a dose of 30 mg/day. given weeks -4 to 0, followed by randomization to double-blinded active Drug well-tolerated. Increased aspartate treatment of lovastatin at 10, 20, 30 or aminotransferase concentrations, with no evidence of a 40 mg/day for 8 weeks. Medication dose-response relationship, and no value exceeding 2x dispensed at weeks -4, 0, and 4. the upper limit of normal. No change in alanine Compliance verified by tablet count at aminotransferase. Three patients had > asymptomatic weeks 0, 4 and 8. Clinic follow-up visits elevations in creatine kinase, which spontaneously at weeks -4, -2, 0, 2, 4, 6, and 8. returned to normal, with no action required regarding the Complete physical exam at weeks 0 drug. and 8. Ophthalmologic exam during placebo period and within 8 weeks of 85% of patients took at least 70% of the prescribed dose study exit. Patients continuously during the active treatment period. counseled by dietician to follow lipid- lowering diet throughout the trial. Daily food records completed by patients for 3 consecutive days and reviewed by a dietician at weeks -4, 0, 4, and 8. 75 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Lambert, 199672 No serious clinical adverse effects reported No separate control group: patients served Fair as own controls 76 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Linder, 1983232 Exercise RCT To determine the effect of 8 weeks Volunteers from middle Augusta, GA an 8-week physical income families, all conditioning program on subjects were members the serum lipid and of Boy Scout troops. lipoprotein levels in healthy Any boys with health white male adolescents. problems and those taking medication were excluded. 77 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Linder, 1983232 50 healthy boys boys ages 11-17; all white randomly assigned to either control (n=21) or physical conditioning (n=29). 78 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Linder, 1983232 Subjects in the physical conditioning 1. 48 hour dietary history 1. Physical working capacity increased in the PA group group were required to participate in 4 kept by parents and from pre to post (16.2+3.4 to 18.7+3.5 kpm/min/kg) as days/week of exercise. This program subjects compared to the control group (16.4+3.3 to 16.3+2.9 consisted of 3 days of an alternating 2. Height, weight, triceps kpm/min/kg). p<0.05 for the difference between the PA walk-jog program (60 seconds each) skin fold, resting pulse and control groups post-test results. and 1 day of competitive team soccer or rate, blood pressure, 2. No significant differences were noted in weight, rugby (min 1 hour). Jogging segments relative body weight height, skinfold, BP or resting pulse rate, dietary were run at a rate that increased the calculated using Weight- components subject's pulse rate to 80% of the for-Length Index. 3. There were no significant differences in the change in maximum pulse rate reached during 3. serum cholesterol levels from pre to post or between individual working capacity testing at groups. baseline. Total cholesterol: PA group 148.14+22.39 pre to Subjects in the control group 147.23+26.49 post vs control group 149.45+24.87 pre to participated in their usual summer 148.22+29.34 post mg/dl activities. LDL: PA group 93.90+20.93 to 97.25+22.86 vs control group 95.20+23.36 to 101.00+33.44 mg/dL HDL: PA group 41.33+8.39 to 37.38+7.87 vs. control group 40.20+8.39 vs 37.94+8.59 VLDL: PA group 15.38+8.15 to 14.47+8.28 vs control group 14.55+10.77 vs 13.89+10.68 Triglycerides: PA group 84.14+29.47 to 84.00+25.40 vs control group 81.70+28.71 vs 87.17+34.12 79 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Linder, 1983232 NR No sessions were missed due to illness; any Fair sessions missed for vacation were completed under parental supervision; 1 day per week was offered as a make-up day. 39 of 50 subjects completed the post-tests; 8 of the 29 subjects in the physical conditioning group were excluded because they failed to fulfill adequately all the requirements of the exercise program. 3 of the 21 subjects in the control group failed to return for follow up testing and were not included in the final analysis. 80 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Malloy, 197873 Drug Single-blind random To determine the effect of 6 months Children aged 5 to 21 assignment pattern p -aminosalicylic (PAS) on each years with severe with cross-over at 6 serum cholesterol and condition familial months - drug triglycerides in 3 (drug and hypercholesterolemia; USA lipoprotein classes in placebo); asting triglycerides were children and adolescents follow-up at normal and with severe familial 12, 18, and electrophoresis of hypercholesterolemia. 24 months serum in agarose gel for some. 4 showed a predominant month run-in narrow liporptien band period for all of beta moility in all prior to patients. randomizatio n. 81 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Malloy, 197873 1. Severe familial hypercholesterolemia 20 enrolled Median age: 12 years 2. Consistently elevated serum cholesterol Female gender: 45% while on > 4 month diet restricted in cholesterol and saturated fats 3. No disorders known to contribute to secondary hyperlipidemia 4. No other serious disorder 5. No medications other than PAS 6. One parent with proven and one other in kindred with hypercholesterolemia. 82 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Malloy, 197873 Dietary cholesterol and saturated fats 1. Serum cholesterol 1. Mean serum cholesterol levels for all subjects were restricted 4 months prior to and 2. Triglycerides 14% lower during the 6 months of treatment with diet throughout the study. Subjects and their 3. LDL and PAS-C vs diet alone (p<.001).Those in the parents received monthly dietary 4. HDL compliant subgroup decreased 19.5% below control counseling. Single-blind assignment to 5. TSH levels with PAS-C (p<.001) active drug (150 mg/kg/day, up to 8 2. Mean fasting serum triglyceride levels of all patients gm/day) or placebo during first 6 were within the normal range with diet alone, and months. Dose adjusted monthly for decreased 15.7% when PAS-C was added (p<.001). change in body weight. After 6 months, Levels in the compliant subgroup were not lower than the alternate agent (placebo or active those of other patients, suggesting a smaller dose of drug) was given for 6 months. PAS-C may be maximally effective. Composite index of compliance for 3. In paired samples of serum from 8 subjects, LDL each subject based on interviews, tacit decreased in all and the mean content of serum drug inventories, and tests for PAS cholesterol in the LDL fraction declined 24% below metabolites. Monthly clinic visits control levels during treatment with drug (p<.002). including clinical exam. 4. No difference in HDL with drug 5. No difference in TSH with drug 83 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Malloy, 197873 No indicators of drug toxicity reported, other than mild gastric Single-blind randomization Poor irritation that remitted with oral antacid treatment. No separate control group: patients served as own controls Only 11/20 subjects were > 50% compliant LDL change evaluated for 8 subjects only 84 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting McCrindle, 199774 Drug Randomized To compare acceptability, 28 weeks 2 pediatric lipid clinics; crossover compliance and children with familial patients recruited effectiveness of two forms hypercholesterolemia from 2 lipid disorder of cholestyramine resin in clinics treatment of hypercholesterolemia in children. 85 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics McCrindle, 199774 1. Age 10 - 18 years 40 43% female 2. 1 or more parent with heterozygous familial 2 siblings from 6 families hypercholesterolemia type IIA or IIB 3 siblings from 1 family 3. Fasting serum LDL of 130 mg/dl or more Median age at enrollment: 13 while on American Heart Association step 2 (10-18) diet Median age at diagnosis: 9.5 3. No contraindications to use of (0.1-17.2) cholestyramine 4. No lipid-lowering agents other than those in study protocol 86 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results McCrindle, 199774 6 week washout period followed by Baseline assessment: self-Reduction in LDL for both pills (-10% +/- 20%, p=0.006) random assignment in pairs to either 8 administered and powder (-15% +/- 17%, p=0.0001) with no significant grams/day of cholestyramine powder (4 questionnaire on difference between forms (p=0.16). gram packets) or pills (1 gram tablets) attitudes, health beliefs At end of study, 82% preferred pill, 16% powder, and 2% for an 8 week period. Drug dispensed and perceptions re: neither form. Mean (+/-SD) compliance as assessed by with random over count, and subjects hypercholesterolemia, amount of medication taken was greater for pills (61% +/- instructed to return unused drug at end detailed medical and 31% than powder (50% +/- 30%, p=0.01). Form of of drug period. Subject completed log family history, food medication increased compliance by at least 25% for 16 book during the 1st week on amount of frequency questionnaire. patients (42%), 13 in favor of pills and 3 of powder. drug taken, acceptability and adverse Fasting lipid profile Compliance not associated with patient attitudes or effects, with questionnaire at weeks 4 Complete blood cell count perceptions of hypercholesterolemia, demographics, and 8. After the first 8 week course of Serum chemistry family history, previous experience with lipid-lowering medication and assessment, there was determinations medications or lipid-profile parameters. another 6 week wash-out, followed by Physical exam crossover to the alternate form of Dietary assessment medication and assessment. Questionnaire on ease of drug use and compliance 87 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating McCrindle, 199774 38/40 (95%) completed both medication periods No control group Fair* Authors noted minor gastrointestinal complaints were frequent Small sample size but did not result in any drop-out At baseline, pill group had perception of increased importance of diet 88 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting McCrindle, 1998235 Diet Randomized trial To determine compliance, 8 weeks Lipid clinic, Toronto, safety, and efficacy of Canada therapy with garlic extract in lowering cholesterol levels in children and adolescents with hypercholesterolemia 89 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics McCrindle, 1998235 Children between ages 8-18 30 randomized to either 16 males, 14 females Positive family history of garlic supplement or Mean age 14.0+2.3 years hypercholesterolemia or prematrues identical placebo atherosclerotic cardiovascular disease in first- degree relative Minimum fasting TC >185 mg/dL Participating in a dietary counseling program Compliance with NCEP Step II diet for at least 6 months Excluded: secondary dyslipidemias, hisotry of major surgery or illness within 3 months of enrollment 90 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results McCrindle, 1998235 1 whole 300 mg tablet of garlic extract Compliance Relative to placebo, garlic treatment resulted in: (Kwai, Lichtwer Pharma, Berlin, Fasting TC, LDL-C, HDL- TC increase 0.6 mmol/L (p=0.86) Germany) containing 0.6 mg of allicin C, TG, and Apo A-1, Apo LDL-C deccrease 0.5mmol/L, p=0.90 placed in a gelatin capsule with inert B-100 HDL increase 9.3 mmol/L, p=0.29 filler. One bulb of Chinese-grown garlic TG decrease 0.72 mmol/L, p=0.70 provides the same amount of allicin as that provided by 6 tablets of garlic extract. Placebo consisted of identical gelatin capsule with inert filler only. 91 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating McCrindle, 1998235 At the end of study, 13% of placebo group and 21% of garlic Fair group reported adverse effect (p=0.66), the most common being headache. Upset stomach was also reported by both groups at the mid-oint questionnaire. No effects on hight, wieght or blood pressure. Serum albumin level and hemoglobin levels increased in the garlic treated group (p=0.002 for albumin and p=0.02 for hemoglobin) Compliance 78+22% of expected for placebo and 72+21% for garlic group. 92 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting McCrindle, 200275 Drug Randomized cross- To determine the short- 18 weeks Lipid clinic, Toronto, over trial term safety and each period Canada effectiveness of combination drug therapy in children McCrindle, 200376 Drug RCT of drug To determine the safety 26 weeks of Children aged 10 to 17 20 centers world- and efficacy of atorvastatin treatment or with FH or severe wide (US, Canada, (10 to 20 mg) in children placebo, then hypercholesterolemia Europe, South and adolescents with all subjects Africa); open label. familial recieved 26 hypercholesterolemia (FH) weeks of or severe treatment hypercholesterolemia 93 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics McCrindle, 200275 1. Children between ages 8-18 40 randomized to either 11 girls, 25 boys 2. Positive family history of colestipol 10grams/day median age 14, range 9-18 hypercholesterolemia or premature alone or colestipol 5 BMI 22-25 kg/m2 atherosclerotic cardiovascular disease in first- grams/day +pravastatin degree relatives 10 mg/day 3. Minimum fasting LDL cholesterol level before enrollment > 160.5 mg/dL 4. Participation and compliance in dietary counseling program for at least 6 months. 5. No secondary cause for hyperlipidemia, no major surgery or serious illness within past 3 months. McCrindle, 200376 1. Known FH or severe hypercholesterolemia 187 Atorvastatin group and LDL-C>190mg/dL or LDL-C > 160 mg/dL randomized, Mean age: 14.1 (2.0) and a family history of FH or documented 140 to drug Female gender: 32% premature CV disease in a 1st or 2nd degree 47 to placebo Race: W: 94% relative; 2. TG < 400 mg/dL; 3. > Tanner B: 1.4% stage II; 4. Not premenarche, pregnant or A: 1.4% breastfeeding; 5. Testicular volume > 3cm3 after age 12; 6. Weight between 10th and Placebo group 95th percentile for age; 7. No liver or kidney Mean age: 14.1 (2.2) disease; 8. No known sensitivity to statins Female gender: 28% Race: W: 87% B: 2.4% A: 2.4% 94 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results McCrindle, 200275 colestipol 10grams/day (10 tablets) Acceptability No data reported prior to cross-over. alone vs colestipol 5 grams/day (5 Compliance tablets) plus pravastatin 10 mg/day (1 LDL, total cholesterol, Results as mean change +SD in mg/dL tablet) for first 18 week period, followed Apo-B 100, triglycerides Total cholesterol: -0.63+0.80 for C vs -1.06+1.10 for by 8 week wash-out period and then C+P, p=0.041 each group crossed over to the other LDL: -0.65+.80 for C vs -1.07+1.06 for C+P, p=0.066 treatment for the second 18 week HDL: +0.01+0.18 for colestipol vs +0.03+0.13 for C+P, period. p=0.63 Triglycerides +0.11+0.68 for colestipol vs -0.07+0.72 for C+P, p=0.28 Apo A-1 (g/L) +0.06+0.16 for C vs +0.07+0.13, p=0.81 Apo B-100 (g/L): -0.19+0.24 for C vs -0.26+0.31 for C+P, p=0.48 McCrindle, 200376 After a 4 week baseline/placebo phase 1. Change in LDL-C, total 1. LDL-C reduced compared to placebo (-40% vs -0.4%, in which subjects were instructed to cholesterol, triglycerides, p<.001). At week 26, change in total cholesterol greater follow the NCEP step 1 diet, those HDL, and apolipoprotein in drug group than in placebo (-32% vs -1.5%; p<.001), whose LDL-C remained > 160 mg/dL B from baseline to week as well as changes in triglycerides (-12% vs + 1.0%; and had TG < 400 mg/dL at week -2 26 p=0.03) and apolipoprotein B (-34% vs +0.7%; p<.001). were randomly assigned in 3:1 ratio to 2. Safety as measured receive 26 weeks double blind through week 52 by HDL cholesterol decreased with drug vs placebo (+2.8% treatment with drug (10mg/d) or standard blood vs -1.8%; p=.02). placebo. Drug could be titrated to 20 measurements, safety mg/d at week 4 for those not at LDL-C laboratories, blood 2. Drug tolerated as well as placebo. Incidence of < 130 mg/dL. Those completing double- pressure, pulse and treatment-related adverse events during double-blind: blind were eligible to continue treatment physical examination 7% in drug vs. 4% in placebo group (p=.7) Most adverse for 26 weeks with open label effects were mild or moderate. Drug had no effect on atorvastatin (10 mg/d). sexual development. 95 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating McCrindle, 200275 Symptoms reported at the end of the study in the final Poor preference questionnaire showed that the majority of patients had no symptoms. 18% reported constipation with colestipol vs 0% with the combination vs 3% who had it with both, 12% reported bloating/gas with colestipol alone vs 0% with combination (3% had it with both). 21% had stomach ache with colestipol alone vs 0% with combination vs 0% with both. 11% had headache with colestipol vs 0% with combination vs. 3% with both regimens. 6% had muscle aches on colestipol along vs 3% with combination vs 0% with both regimens. McCrindle, 200376 No significant adverse effects documented Good 96 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Obarzanek, 2001218 Diet RCT To evaluate the long-term 4 years; Prepubertal girls aged 6 clinical centers in safety and efficacy of a treatment 7.8 - 10.1 years and the USA cholesterol-lowering diet in changed at 6 boys aged 8.6 - 10.8 children with elevated LDL months, 12 years were recruited months, and from schools, an HMO, at 2 and 3 and pediatric practices. years. 97 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Obarzanek, 2001218 1. Average of 2 LDL-C values in 80th - 98th 663 randomized, Intervention group sex-specific percentiles for 8 - 10 year old 334 to intervention Mean age: 9.5 (.74) children in a fasting state 329 to control Female gender: 46% 2. No medical conditions or medications that Race: 86.5% of entire study could effect growth or serum cholesterol, sample white behavioral problems or onset of pubertal maturation. Control group Mean age: 9.5 (.70) Female gender: 44% Race: 86.5% of entire study sample white 98 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Obarzanek, 2001218 Randomized to dietary intervention or 1. Dietary intake 1. Reductions in dietary total fat, saturated fat were control group. 2. Serum LDL-C, height, greater in intervention than in control throughout the serum ferritin intervention period (p<.001). Differences in dietary Intervention: Dietary behavioral 3. Serum total cholesterol, cholesterol intake between intervention and usual care intervention that promoted adherence to HDL-C, LDL-C/HDL-C, groups were significant at 1, 3, and 5 years (all p<.001), a diet with 28% of energy from total triglycerides, red blood but not at the last visit. At 1 and 3 years energy intake fat,< 8% from saturated fat, up to 9% cell folate, serum retinol was lower in intervention vs control (p=.01 and p<.0001, from polyunsaturated fat, and <75 and zinc, sexual respectively) mg/1000 kcal cholesterol per day. maturation 2. At 1, 3, and 7 years follow-up, intervention had 4.8 mg/dL, 3.3 mg/dL, and 2.0 mg/dL lower LDL-C than Control: Usual care - parents informed control, respectively. Mean adjusted differences between child's blood cholesterol high and were the 2 groups were significant at 1 and 3 years (p<.001 given educational materials on heart- and p<.02, respectively), but not at 5 years (p=.11) or healthy eating as available to the public. the last visit (p=.25). No significant differences in height Annual examinations for study-wide or serum ferritin. measurements. 3. Total serum cholesterol lower in intervention than Intervention: 1st six months: 6 weekly usual care at 1 (p<.001) and 3 years (p=.04), but not at 5 and then 5 biweekly group sessions led or 7 years. HDL-C was higher in control than intervention by nutritionists and behaviorists, and 2 at 1 year (p=.03), with no difference at years 3, 5 or 7. individual sessions with nutritionist. 2nd No differences seen in LDL-C/HDL-C or triglycerides. six months: 4 group and 2 individual Serum retinol was higher in the intervention group at 3 (p= sessions. 2nd and 3rd years: group and 7(p=.02) years. There were no other differences between individual maintenance sessions held 4 - groups in nutritional biochemical measures. There were n 6 x/year, with monthly phone contacts differences in weight or measures of sexual maturation, o between group sessions. 4th year: intake of recommended vitamins and minerals. individualized approach with motivational interviewing and stage of change. 2 group events and 2 individual visits annually, with individual phone contacts "as appropriate". 99 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Obarzanek, 2001218 No differences at any data collection point in height or serum Higher proportion of intervention vs control Good ferritin or any differences in an adverse direction in red blood had household income < $20,000 (15.1% cell folate, serum retinol and zinc, sexual maturation or body vs. 5.9%; p=.002). Secondary analyses mass index. suggests this may have effected results. 87.5% participated in final follow-up Only 1.3 of 4 scheduled intervention visits per subject were achieved in the final year. Intention to treat analysis conducted Dietary intake based on self-report and may be biased. 100 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Savage, 1986233 Exercise RCT To compare the serum 11 weeks Subjects recruited by Omaha, Nebraska lipid and lipoprotein media from Omaha, response to identical Nebraska metropolitan exercise training in young area children and adults. 101 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Savage, 1986233 Inclusion criteria: Subjects were randomly Ages for boys not listed, all 1. Prepubescent boys (no pubic hair as divided in to high, low prepubescent confirmed by parents) or adult males ages 30- and control groups (n=8 45. for Low, n=12 for High, 2. No regular (>3x/week) aerobic exercise n=10 for control). program within the last 3 months. 3. No medical contraindications 4. Medical examination and clearance by a physician within the last year. 5. Non-smoker. 102 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Savage, 1986233 Training consisted of 1. Serum lipids For boys only: results as pre-post for low (L), high (H) walking/jogging/running 3 times/week 2. Cardiorespiratory treatment groups and control (C) mean+sd. for 11 week for a total distance of 1.6 fitness (modified Bruce km per session. Participants had a 1 ETT), VO2 max. HDL/cholesterol (%): Low group decreased 36+3.9 to week period of accommodation 3. Body composition: 28+2.8; High group decreased 40+1.4 to 32+2.0; Control preceding the training period. Low weight, height, skinfold group 32+2.0 to 32+2.3. NS intensity group had a training HR of thickness, body density Triglycerides (mg/dL): Low group increased 50+2.1 to 40% VO2 max. High intensity training 4. Dietary intake for 3 55+3.5; High group increased 53+3.6 to 69+8.6; Control group had a training HR of 75% VO2 days. group increased 68+8.1 to 81+5.8. NS max. Control subjects were requested LDL (mg/dL): Low group increased 109+20.0 to to maintain their current activity pattern 123+23.2; High group increased 88+3.8 to 92+3.6; throughout the study. Control group decreased 96+7.8 to 93+8.8. NS All post values were the average of two measurements taken on day 1 and day 2 post-completion of the training regimen. No significantly differences at p=0.05 by ANCOVA. 103 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Savage, 1986233 NR Fair 104 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Shannon, 1994220 Diet RCT To assess effects of a 3 month Children aged 4 -10 with home-based, parent-child follow-up elevated plasma LDL-C autotutorial (PCAT) dietary identified through a education program on the cholesterol screening dietary knowledge, lipid program open to those consumption, and plasma served in 9 pediatric LDL of 4 to 10 year old practices in suburban children with elevated Philadelphia. plasma LDL-C. USA 105 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Shannon, 1994220 1. Age 3.9 - 9.9 years 342 Average age: 6.3 2. 85% to 130% ideal body weight for height Girls: 50% 3. No history of disease that would explain Boys: 50% hypercholesterolemia White: 76.8% 4. No medications that would affect blood African-American: 6.75% lipids Other: 2% 5. Boys: mean fasting plasma LDL-C 107 - 164 mg/dL. Girls: 112 - 164 mg/dL. Not at risk control group: 1 through 4, and TC at screening test below 60th percentile 106 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Shannon, 1994220 Randomization into(1) Parent-child Dietary knowledge Dietary knowledge scores: increased 3x more in PCAT autotutorial program (PCAT) (n=88), (2) Lipid intake vs counseling or at-risk control (p<.001). Lipid intake: individual dietary counseling (n=86) or Plasma LDL-C level Mean grams of total and saturated fat decreased in (3) at-risk-control groups (n=87) using Weight and height PCAT and counseling and increased slightly in at-risk permuted blocks within strata design control (p<.05). At-risk control knowledge and diet did with an adaptive allocation procedure to not differ from that of not-at-risk group. Mean LDL-C balance 1st order interactions with decline of PCAT (10.1mg/dL) was greater than in at-risk season and pediatric practice. control (4.1 mg/dL) (p<.05). This represents an 8% Adjustment made to select a decline in the PCAT group. comparable control group among children "not at risk" (n=81). After initial assessment, PCAT treatment included 10 talking book lessons and follow-up paper and pencil games for children with a manual for parents. Counseling treatment: 45-60 minute session with parent, child and registered dietitian, along with take home print materials for both. Dietitian available for phone to answer questions during 3 month period. 107 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Shannon, 1994220 Not Reported At baseline, mean calorie intake of at-risk Good control group greater than in other groups (p<.05). 108 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Stallings, 199377 Diet Randomized pilot To determine the 6 months. Healthy children aged 4 study effectiveness of a home- PCAT is a 10-to 10 with high LDL USA based dietary education week lesson cholesterol referred to a program vs usual care in program; research center. LDL, gain in knowledge of families got heart-healthy eating, and that and change in consumption of nothing else dietary fat for 6 months. Usual care families got 1 visit with a dietician and nothing else for 6 monhts. Assessments (serum and diet) at 3 and 6 months. 109 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Stallings, 199377 1. Plasma LDL between 90th - 99th 44 randomized, 20 to PCAT group percentile PCAT Mean age: 8.3 (2.5) 24 to usual care Female gender: 40% Race: NR Placebo group Mean age: 7.7 (2.2) Female gender: 42% Race: NR 110 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Stallings, 199377 After assessment and instruction to 1. Change from baseline 1. No differences between groups at each time period; parents on use of measurements, in LDL cholesterol within both groups 3 and 6 month values were different randomized to PCAT or Usual Care. response from baseline (p<.01). Both groups maintained a PCAT: 10 lessons, 1 per week 2. Gain in knowledge of decrease of about 10% over 6 months. completed in home by child and heart healthy eating 2. Knowledge scores of the 2 groups did not differ parents. Registered pediatric dietitian patterns during the study. Both groups had increases in explained use of the program for self- 3. Change in consumption knowledge at 3 and 6 months compared to baseline (p< education. patterns of dietary fats 0.001). Usual care: 1 hour session with 3. Calories: No difference between or within groups at registered pediatric dietitian who each time period. provided standardized guidance using Total fat: No difference between groups at each time an American Heart Association booklet. period. Within both groups, 6 month values different from Assessment at 3 and 6 months baseline (p<0.05). Saturated fat: No difference between or within groups at each time period. Cholesterol: At 6 months, PCAT lower than Usual Care (p<0.05). Within both groups, 3 month values lower than baseline (p<0.05). At 6 month follow-up, Usual Care higher than at baseline (160 vs 89). 111 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Stallings, 199377 Not Reported 86% completed 3 month follow-up. Poor 68% completed 6 month follow-up. Drop -out in specific variable data-sets was higher - EG: 66% completed knowledge test at 6 months. High drop-out rate with resulting small N's in many comparisons. Dietary data based on self-report. 112 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Stein, 199978 Drug RCT of drug To assess the lipid- 48 weeks Boys aged 10 - 17 years 14 pediatric clinics in lowering efficacy, with HeFH recruited the USA and Finland biochemical safety, and from 14 pediatric effect on growth and outpatient clinics sexual development of lovastatin in adolescent boys with HeFH 113 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Stein, 199978 1. Followed American Heart Association 132 randomized, Lovastatin group (AHA) pediatric diet for at least 4 months 67 to drug Mean age: 13.3 (0.3) 2. LDL-C 189 mg/dL - 503 mg/dL 65 placebo Female gender: 0% 3. At least 1 parent had an LDL-C of at least Race: 93% of all were white 189 mg/dL not due to secondary causes or LDL-C values were 220 - 503 mg/dL or their Placebo group LDL-C values were 220 - 503 mg/dL and a Mean age: 13.1 (0.3) parent died of CAD with no available lipid Female gender: 0% values. Race: 93% of all were white 4. No homozygous FH, underlying disorders known to produce secondary LDL-C elevations 5. No disorders affecting triglyceride-rich lipoproteins 6. After the trial began, FDA requested only subjects who had Tanner stage 2 at entry be included in the trial. 7. No delayed puberty shown by testicular volume < 3 cm after age 12 8. Weight > 32 kg and between 10th - 95th percentile for age 114 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Stein, 199978 All received formal dietary reinstruction, 1. LDL-C 1. Lovastatin: reductions in LDL-C and total cholesterol were monitored, and determined to be 2. Growth at all dosages (p<.001 vs placebo) and in apoB at 40 stable following a AHA pediatric diet for 3. Sexual development mg/day. About 6% additional LDL-C lowering occurred at least 8 weeks before randomization. 4. Biochemical, nutritional with each doubling of lovastatin dosage. No significant Placebo was administered week -4 to and endocrine changes in HDL-C or triglycerides compared to placebo. week 0. All subjects continued diet parameters No changes in either group in Lp (a). During weeks 24- instruction, monitoring, and evaluation 48, lovastatin reduced LDL-C and apo B 25% and 22%, throughout the study. respectively. No significant changes in other lipids or Randomized to drug: Lovastatin, apoliproteins. starting at 10 mg/day, with a forced 2. Drug had no significant effect on growth parameters at titration at 8 and 16 weeks to 20 and 40 24 or 48 weeks mg/day, respectively. 40 mg/day weeks 3. No difference between drug and placebo in testicular 25 - 48. volume, Tanner staging, testosterone, or LH level. Randomized to placebo: Matching DHEAS increased 18% in the drug group compared to placebo tablets 5% in the placebo group. (p=.03) 4. Reduction of tocopherol in drug group at week 48 (p=.002), consistent with decrease in LDL-C. In both groups, increase in ALT from baseline, with no differences between groups at week 48 (p=.20), although increase was significant for each group (drug, P<.001; placebo, P=.008). No consistent changes in AST or CK. Report of infrequent, sporadic, nonsustained CK elevations in response to exercise. 115 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Stein, 199978 1 drug and 2 placebo subjects discontinued due to adverse Good effects. No clinically significant adverse effects. 116 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Stergioulas, Exercise RCT To investigate how 2 months Sedentary high school 1998228 Greece physical exercise on a boys aged 10 - 14 years bicycle ergometer recruited from the influences serum high- random population density lipoprotein study sample examined cholesterol and in the survey of the prostacyclin levels Greek Secretariat of Sports in 1993. 117 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Stergioulas, 1. Resident of Peania 58 randomized, Exercise group 1998228 2. Lipids and prostanoids between normal 38 to exercise Mean age: 11.9 (0.5) concentrations (Szamosi et al) 20 to control Female gender: 0% 3. Body weight between normal values (8 -10 Race: kg lower the height) Control group Mean age: 12.1 (0.6) Female gender: 0% Race: 118 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Stergioulas, 4 month lead-in phase. Randomized to 1. Physical work capacity 1. Physical work capacity of exercise group increased at 1998228 exercise intervention or control. 2. HDL-C the end of treatment (p<0.001). No change seen in 3. 6-keto-PGF1a control group. Exercise: 2 month training program with concentration 2. Exercise group had higher HDL-C at end of the 2nd four, 60 minute sessions per week. treatment month (p<0.005) and at the end of the Sessions included warm-up, stretching, detraining period (p<0.01). No change seen in control and aerobic exercise at 75% of physical group. work capacity. 3. Exercise group had higher 6-keto-PGF1a concentration at end of 2nd treatment month compared Control: No specific training program. to baseline (p<0.001) and lower 6-keto-PGF1a concentration at end of the detraining period compared Both groups advised not to change to the end of the 2nd treatment month (p<0.001). No dietary habits or physical activity. Blood change seen in control group. samples taken at baseline, after 15 days of training, after 1st and 2nd training month, at end of 1st month of detraining. 119 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Stergioulas, None reported. Very high attrition/drop-out rate: Poor 1998228 intervention 20/38 (53%) dropped control 10/20 (50%) dropped. Detraining period not defined. Four month lead-in phase not described. No analyses of between group differences. 120 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Tonstad, 199680 Drug Double-blind RCT of Effectiveness of colestipol 67 adolescents colestipol 10g once in treating FH. Norway per day or 5 g twice per day versus placebo 121 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Tonstad, 199680 Adolescents age 10-16 referred to lipid clinic; 74 targeted, 67 agreed to 37 boys (58%); 29 girls all had two or more TC >239.75 mg/dL and participate mean age 13.2 TG<116.01 mg/dL. Tanner stage 2.6 122 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Tonstad, 199680 All subjects instructed to follow a diet: 1. Fasting TC, TG, HDL, After 8 weeks: 30% calories from fat; <10% from calculated LDL, 1. TC decreased by 14% in colestipol ( 8.16+1.47 pre saturated fat and <200 mg apolipoproteins A-I and B, vs. 6.99+1.46 post) vs 1% in placebo (7.69+1.26 pre vs cholesterol/day. Assessed with 4 day serum carotenoids. 7.57+1.34 post), p<0.01; weighed food record. 2. Compliance LDL decreased by 19.5% in colestipol (6.58+1.3 pre vs Subjects randomized to drug treatment 3. Diet 5.27+1.29 post) vs 1% in placebo (6.13+1.18 pre vs. got either colestipol 10g once per day 4. Growth 6.01+1.32 post), p<0.01. or 5g twice per day x 8 weeks. HDL was not significantly different (colestipol 1.12+0.25 Open label trial after 8 weeks in which pre vs 1.19+0.35 post vs placebo 1.12+0.20 vs placebo group started colestipol for 52 1.18+0.23). weeks; colestipol group continued on TG were not significantly different (colestipol colestipol. 0.99+0.61pre; 1.15+0.90 post vs placebo 0.96+0.66 pre vs 0.84+0.36 post). 2. Compliance was 68% in colestipol group vs 76% in placebo group NS. 3. intakes of major nutrients and vitamins were similar in the groups 4. Linear growth velocity was within reference range for Norwegian adolescents. 1 patient lost >1 kg during the study (24.5 kg/m2 pre and 21.3 kg/m2 post). 123 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Tonstad, 199680 Total of 7 subjects dropped out 4 in colestipol; 3 in placebo. 5 placebo group was older and more Poor patients dropped out because they refused to swallow the drug; advanced Tanner stage (p<0.05) 2 dropped out because they had unexpectedly low lipid concentrations at randomization. 8 girls did not agree to continue in the open phase. Of 51 in open phase, 42 completed 1 year. 8 subjects in colestipol group had GI side effects attributed to the drug: 2 had constipation that improved with temporary dose reduction; 1 had dyspepsia and 1 had flatulence throughout entire study; 2 had intermittent nausea, 1 had temporary decrease in appetite; 1 had abdominal pain improved with dose reduction. 124 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Tonstad, 199681 Drug Random assignment To determine the efficacy 2 years: 1 Boys aged 6 to 11 years to double-blind and safety of year of diet and girls aged 6 to 10 comparison cholestyramine therapy in for everyone; years with familial Norway young children with familial 1 year hypercholesterolemia in hypercholesterolemia randomized a referral lipid clinic treatment phase 125 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Tonstad, 199681 1. Familial hypercholesterolemia diagnosed if 96 randomized, Drug group TC > 260 mg/dl and triglyceride levels < 200 36 to drug Mean age: NR mg/dl, if 1 parent had baseline cholesterol > 36 to placebo Female gender: 44% 300 mg/dl and triglyceride levels < 200 mg/dl, Race: NR or tendon xanthoma, or if autosomal dominant inheritance present in other Placebo group members. Mean age: NR 2. Prepubertal girls aged 6 - 10, and Female gender: 33% prepubertal boys aged 6 - 11 who completed Race: NR an initial medical evaluation and dietary session and had a parent who understood the diet. 4. Height velocity score standard deviation score > -2.0. 3. Those taking lipid-lowering drugs were required to discontinue doing so at least 4 weeks before study start 4. Secondary hyperlipidemia ruled out by clinical and lab examination. 126 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Tonstad, 199681 After 1 year of a low-fat, low-cholesterol 1. Serum LDL 1. Serum cholesterol decreased in drug group vs diet, children with LDL > 190 mg/dl or > 2. Height velocity placebo after 2, 9 and 12 months (p<0.001) LDL 160 mg/dl in the presence of familial 3. Erythrocyte folate decreased by -16.9% to -18.6% vs 0 to 1.5% (p=0.0001). premature cardiovascular disease were 4. Total plasma Change in HDL level was no significant, and mean randomly assigned to a double-blind homocysteine triglyceride levels remained unchanged in both groups. comparison of 8 gm cholestyramine or 5. Serum fat-soluble Apolipoprotein B levels were reduced in drug vs placebo placebo for 1 year. Dieticians reinforced vitamins groups (p=0.0001). the diet at each visit. Dietary 6. Side-effects 2. No difference in mean height velocity standard composition assessed by a 7 day deviation scores in those who hadn't started puberty record, evaluated by FIBER. Assigned 3. No significant differences reported dose 8 gm/day, following 1 week build- 4. Total homocysteine level increased in drug group, and up phase of 4 gm/day. Follow-up visits was negatively correlated with levels of erythrocyte and scheduled after 2, 5, 7, 9 and 12 serum folate after 1 year. months. Compliance assessed by 5. In drug group vs placebo, mean levels of 25- counting leftover packages at each visit. hydroxyvitamin D decreased (p=0.04) 6. In drug group, one girl had low folate and elevated homocysteine levels, and there was one case of intestinal obstruction caused by adhesions 127 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Tonstad, 199681 In the drug group, folate defiency occurred in one subject, in 1. High attrition: only 22/36 (61%) in drug Fair whom the total homocysteine level increased concurrently. With group and 26/36 (72%) in placebo the exception of modest decreases in levels of vitamin D during completed the study. the winter, other nutritional deficiencies were not observed. 2. Only about half the children were able to No adverse effects on growth and bone age evident. take cholestyramine in the amount prescribed, primarily because of unpalatability. 128 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Walter, 1985222 Diet and Randomized trials of To assess the feasibility 5 years 2,283 4th graders from exercise school-based and effectiveness of a 22 elementary schools combination interventions curriculum focused on in the Bronx, NY. nutrition, physical fitness, and smoking prevention in elementary schools 129 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Walter, 1985222 Assume all children at each school were 1,563 (68.5% mean age=9.1 years eligible participated in baseline male 51.4% examination. 1,115 of race: these (71.3% )were black 48.9% available at 1 year for Hispanic 23.2% follow up assessments. white 24.6% Final n: 805 intervention 3.3% other (Asian or Pacific and 310 control origin primarily) median family income $22,126 130 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Walter, 1985222 Intervention schools received the 1. HDL, TC 1. Plasma TC: control (170.1 mg/ml+25.6 at baseline "Know Your Body" curriculum each year 2. Serum thiocyanate vs 171+28.3 at 1 year; intervention 173.3+26.6 at BL vs starting in 4th grade and continuing until ponderosity index 172.7+27.7 at year 1). This is a 0.4% decrease in the 8th grade; taught in usual classroom by (derived from height and intervention group compared with 0.5% increase in regular teacher 2 hours/week for entire weight measurements) control group. school year. 3. Triceps skinfold 2. Plasma HDL: control 56.8+12.8 at BL vs 60.5+12.3 Teachers trained by research staff in 3 thickness at 1 year; intervention 54.4+11.3 at BL vs 59.5+13.5 at 1 half-day teacher workshops on 4. Post-exercise pulse year. curriculum implementation. recovery rate (recovery 3. TC/HDL ratio control 3.1+0.8 at BL vs 2.9+0.7 at 1 Adherence to teaching protocols index). year; intervention 3.3+1.2 at BL vs 3.0+0.8 at 1 year. monitored by classroom visits from 5. Systolic and diastolic research staff and attendance at BP Adjusting for age, gender, race and BL risk factor level, training workshops. TC difference between Intervention and Control groups Curriculum based on social learning, at 1 year was -3.1 (p=0.032); HDL difference was +1.0 encouraging adoption of healthy (p=0.175); ratio difference was -0.10 (p=-0.063). behaviors: AHA prudent diet/AHA Committee Report (1978); regular Intervention was successful in reducing systolic and exercise program; resisting social diastolic BP. pressures to smoke. Curriculum materials included: teacher guides, student workbooks and worksheets, health passports, videotapes, posters and calendars. 131 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Walter, 1985222 Significant impact on the classroom teacher (time away from Screening for risk factors was also done in Fair usual material) control school children. 132 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Wheeler, 198582 Drug DB randomized To evaluate the use of 6 months: 3 14 children with FH cross-over trial benafibrate for lower month arm of cholesterol in children. either placebo or medication. Wiegman, 200437 Drug RCT of drug To determine the 2-year 2 years FH children ages 8 to 18 Netherlands efficacy and safety of recruited from academic pravastatin therapy in medical referral setting children with familial hypercholesterolemia (FH) 133 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Wheeler, 198582 Diagnosis of FH established by TC>259.09 14 children; all had NR mg/dL, a type IIa pattern on lipoprotein previously been treated electrophoresis and normal fasting TG (<58 with dietary measures mg/dL) in the patient; with either similar (inadequate to control lipoprotein abnormalities in one of the parents cholesterol), had been or premature CHD death in a parent and a recommended similar lipid abnormality in another close cholestyramine but relative. refused to take it. Wiegman, 200437 1.Parent with clinical or molecular diagnosis 214 randomized, 106 to Pravastatin group of FH; 2.Two fasting samples with LDH-C drug Mean age: 13.0 levels of 155 mg/dL or more and triglyceride 108 to placebo Female gender: 57% levels below 350 mg/dL; 3.Adequate Race: NR contraception for sexually active girls; 4. No drug treatment for FH or use of plant sterols; Placebo group 5. No homozygous FH; 6. No hypothyroidism; Mean age: 13.0 7. No abnormal levels of muscle or liver Female gender: 57% enzymes. Race: NR 134 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Wheeler, 198582 1. Children asked not to change their 1. TC, HDL, TG 1. Mean plasma TC on bezafibrate was 22% lower than diets 2. Urine bezafibrate to while on placebo (7.8+1.0 vs 10.0+1.6, p<0.0001) and 2. 10-20 mg/kg/day given bid (in 100 assess compliance. 16% lower than in the period before the trial (7.8+1.0 vs. mg tablets) given for 3 months 9.3+1.5,p<0.001). 2. Mean HDL rose 15% while on bezafibrate compared with placebo (p<0.01); increased 25% compared with pre-trial values (p<0.001). 3. Mean plasma TG fell 23% while on bezafibrate compared with placebo (pNS), and 33% compared with pre-trial values (p<0.05). 4. In all but one case bezafibrate was detected in the urine on all measurements during the active drug phase, and in none of the measurements while on placebo. Wiegman, 200437 After initiation of a fat restricted diet for 1. Change from baseline 1. Pravastatin: trend toward regression, mean [SD], - 3 months or more and encouragement mean carotid IMT 0.010 [0.48]mm; p=.049. Placebo: trend toward of regular physical activity, randomized 2. Growth, maturation and progression, 0.005 [0.044]mm;p=.28. to pravastatin 20 - 40 mg/day or hormone level Mean change in IMT between the groups significant, placebo. Evaluated every 6 months for measurements p=.02. Greater reduction in mean LDL with pravastatin 2 years. 3. Changes in muscle and vs placebo (-24.1% vs 0.3% respectively, p<.001. liver enzyme levels 2. No differences observed in growth, endocrine function parameters, Tanner staging scores, onset of menses or testicular volume. 3. No differences observed in muscle or liver enzyme levels. 135 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Wheeler, 198582 1. One child had high alk phosphatase at the end of 3 months All children declared a strong preference for Poor of bezafibrate (had a slight intercurrent infection, and values this drug as compared with cholestyramine returned to normal) used previously. 2. Another child had slight transient rise in alanine No results prior to cross-over. transaminase during first 2 months on bezafibrate but values No wash out period between cross over were normal by end of 3rd month periods. 3. growth was satisfactory throughout trial 4. no other reports of adverse effects Wiegman, 200437 No significant differences between groups for "all endocrine 2 year follow-up Good functionp arameters." In the placebo group, 1 child developed Well-designed extreme CPK elevation which returned to normal 1 week after stopping the study regimen. In the placebo group, AST elevation (more than 3-fold) occured twice. Elevations in CPK occured 4 times in the treatment group and three times in the placebo group. "No adverse effects on growth, sexual maturation, hormone levels, or liver or muscle tissue." 136 Evidence Table 4. RCTS of Treatment Type of Type of Study/ Duration of Author, year Intervention Setting Aims Trial Population/ Setting Williams, 1995229 Diet Randomized single to compare fiber enriched 12 week Children ages 2-11 blind trial of psyllium Step I diet to usual Step I referred to lipid clinic diet to lower cholesterol in children with borderline high and high blood cholesterol levels 137 Evidence Table 4. RCTS of Treatment Author, year Main eligibility criteria Enrolled Demographics Williams, 1995229 Serum cholesterol >170 mg/dL and LDL >110 58 children 5% of children had a parent mg/dL and 52% had a grandparent with premature onset of CHD (<55 yrs); 48% had a parent with blood cholesterol >200 mg/dL 138 Evidence Table 4. RCTS of Treatment Author, year Interventions Outcomes Assessed Results Williams, 1995229 Age 2-11 years (mean age in Step I 1 box of cereal per day x Lipoprotein results n=50; anthropometric and lipid results group 94.7mo (n=24); in fiber group 1st 3 weeks, then 2 N=49. 82.5mo n=26) boxes/day. Children age After 12 weeks: 2-5 consumed only 1 1. TC 196 mg/dL step I (-5.5%) vs 197mg/dL (-9.6%) box/day throughout study. fiber group; p<0.05 between groups, p<0.01 within Step Fiber group cereal I, p<0.001 with fiber. contained 3.2 g soluble 2. HDL 49.6+8.8 mg/dL step I (+1.5%) vs fiber per serving; control 45.01+13.57mg/dL (+4%) fiber group, NS between cereal contained 0.5 g groups or within groups. soluble fiber per serving. 3. LDL 124.3+25 step I (-6.4%) vs. 126.76 (-15.7%); Both identical boxes, p<0.001 between groups, NS change in step I group, manufacturer. p<0.01 for change in fiber group. 3. TG 113.9+47 mg/dL step I (-15.3%) vs 128.9+55mg/dL (-8.9%) fiber group (p<0.001 between groups; NS change within groups) 4. TC/HDL and LDL/HDL ratios also had significant improvements (-17.9% and -21.1% respectively) in the fiber group vs -4.7% and -7.4% in Step I group, P<0.001 for between group difference. 139 Evidence Table 4. RCTS of Treatment Quality Author, year Adverse Effects Comments Rating Williams, 1995229 Growth velocity, height, weight skinfold thickness and BMI were Authors state that 8 children who did not Poor monitored and no differences between treatment and control complete the study were no significantly group were noted. different than the remaining 50. 140 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Statins Clauss, 200638 Lipid-lowering Effects of RCT in Post-menarchal NR 24 weeks Lovastatin in Post-menarchal girls with heFH Girls with heterozygous Familial Hypercholesterolemia Couture, Association of specific LDL RCT To determine whether the nature of the LDL 6 weeks 199866 receptor gene mutations with Laval University Lipid receptor mutation affects the response to differential plasma lipoprotein Research Clinic, Quebec simvastatin. To describe the different responses to response to simvastatin in young City, Canada simvastatin of plasma lipids, lipoproteins, and French Canadians with apoprotein levels among 3 genetically differentiated heterozygous familial groups of heterozygous FH children and hypercholesterolemia adolescents. de Jongh, Early statin therapy restores RCT with an additional To determine whether simvastatins improves 28 weeks 2002a68 endothelial function in children arm of non-FH controls. endothelial function in children with familial with familial hypercholesterolemia USA. hypercholesterolemia (FH) 1 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Statins Clauss, 200638 HeFH; baseline LDL required was 160-400 mg/dL and parental history of 54 girls ages 10-17 at - familial hypercholesterolemia. least 1 years post- menarche Couture, 1. Heterozygous FH patients with 1of 3 mutations in the LDL receptor gene. 63 FH patients at 37 boys and 26 girls 199866 2. Aged 8 - 17 university research clinic 3. Weight > 27 kg 4. Plasma levels persistently above 95th percentile for age and sex while maintaining a lipid-lowering diet. 5. No concomitant conditions, such as diabetes, anorexia, thyroid disorders de Jongh, Children with heterozygous FH as defined by: 1. LDL>95 percentile for 50 heterozygous FH HeFH patients ages 9-18. 2002a68 age/gender; 2. Documented family history of hyperlipidemia with LDL>95 children plus 19 non- FH simvastatin group Mean percentile for age/gender after treatment; 3. Personal diagnosis by detection affected controls. age: 14.6+2 Female of mutation at the LDL receptor gene 28 in FH simvastatin gender: 13 (42%) Race: group, 22 in FH placebo NR. FH placebo group group, 19 non-FH Mean age: 14.6+2.5. controls. Female gender: 11 (50%) Race: NR. 2 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Statins Clauss, 200638 Lovastatin 20 mg for 1st 4 weeks, then 40 mg thereafter "Adverse reactions to lovastatin were generally mild and transient." Patients treated with lovastatin had an adverse experience profile generally similar to that of patients treated with placebo. In this limited controlled study, there was no detectable effect on menstrual cycle length in girls. Adolescent females should be counseled on appropriate contraceptive methods while on lovastatin therapy." Couture, Individual screening with medical history, physical exam, Patients were questioned about any adverse or unusual signs or 199866 interview with dietician, and blood sampling 6 weeks prior to symptoms, but none were suggested. Compliance to the prescribed study entry. All lipid-lowering medications discontinued at medication was assessed weekly by tablet counting, and there was no screening entry. Four weeks placebo run-in, followed by significant difference in compliance among the mutation groups. randomization to double-blind active treatment, 20 mg/d simvastatin or placebo for 6 weeks at 3:1 ratio. Compliance verified by tablet count at weeks 0, 2, 4, and 6. Patients questioned about adverse or unusual signs or symptoms at all clinic follow-up visits (weeks 0, 2, 4, and 6). Patients counseled by dietician to follow American Heart Association phase I diet throughout trial. Routine hematology and blood chemistry test results were obtained at week -6 and 6. de Jongh, 50 heterozygous FH children randomized to simvastatin or There were no significant differences with regard to safety 2002a68 placebo in 3:2 ratio. An additional 19 healthy non-FH siblings measurements (ALT, AST, and CK) between simvastatin and placebo also used as controls. FH groups, and no adverse events were reported. 3 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Statins Clauss, 200638 NR Couture, NR 199866 de Jongh, Not reported. 2002a68 4 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study De Jongh, Efficacy and safety of statin Double-blind, placebo To evaluate LDL-cholesterol-lowering efficacy, 48 weeks of 2002b69 therapy in children with familial controlled multi-center overall safety, tolerability and the influence of growth intervention after 4 hypercholesterolemia RCT and pubertal development of simvastatin in a large week run-in 9 international sites cohort of boys and girls with heterozygous familial hypercholesterolemia (heFH). De Jongh, Quality of life, anxiety and Cross-sectional study; lipid To determine whether statin therapy influenced the 0 (cross-sectional 200370 concerns among statin-treated clinic, Amsterdam quality of life and anxiety levels of the FH children survey) children with familial and their parents, and to assess the specific FH- hypercholesterolemia and their related concerns of these families. parents Dirisamer, The effect of low-dose simvastatin Single-center, open-label, To investigate the efficacy and safety of low-dose 18 months: 200370 in children with familial diet-controlled study. simvastatin for 1 year in children and adolescents 3-month diet, hypercholesterolemia: a 1-year (aged 10-17) with FH. Safety assessment included followed by drug observation weight, height, and lab values. therapy 5 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics De Jongh, 1. Age 10 - 17 173 98 boys and 75 girls 2002b69 2. LDL-C 4.1 - 10.3 mmol/L 3. Parent with confirmed diagnosis of heFH. 4. Boys: Tanner stage II or above Girls: postmenarchal for at least 1 year before study initiation Exclusion: familial hypercholesterolemia, secondary hyperlipidemia De Jongh, Children who had received 1 year of statin therapy and their parents were 69 children from 51 Mean age 15.3 200370 recruited from the Paedriatric Lipid Clinic of the Emma Children's Hospital of families 47.8% female the University of Amsterdam. Inclusion criteria: age 10-18 y, LDL >158 before statin therapy and one parent diagnosed with FH. Dirisamer, Boys and girls aged 10-17 with LDL >190 and a BMI between 5th and 85th 20 Mean age 13.0, range 10-17 200370 percentile for age were included. 60% female 6 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects De Jongh, After 4 week placebo/diet run-in, randomization to active 223 assessed for eligibility, 175 randomized - 69 to placebo, 106 to 2002b69 treatment or matching placebo in 3:2 ratio and stratified by sex. control. 5 of placebo discontinued: 1 lost to follow-up, 2 withdrew Simvastatin (drug) started at 10mg/day and increased at 8 week consent, 2 for other reasons. 5 of drug discontinued: 1 to intervals to 20 and then 40 mg/day for the remainder of the study mononucleosis, 1 to protocol deviation, and 3 to consent withdrawal. (period 1, 24 week duration) and for a 24 week extension (period No serious AEs. Total drug group clinical AEs included: abdominal 2). Office visits every 4 weeks. Menstrual cycle monitored pain (3), chest pain (1), flatulence (1), myalgia (2), headache (4), sleep throughout study period and Tanner staging used to assess disorder (1), weight gain (1), pruritus (1). Total drug group lab AEs: pubertal development. increased ALT (3), increased AST (3), increased CK (1). No deleterious effects on growth or pubertal development. De Jongh, Children had been treated with statins for 1 year prior to the There were no significant differences between the children with FH and 200370 study. The children and their parents were administered their healthy peers on the health-related quality of life and anxiety questionnaires during a visit to the Lipid Clinic while an questionnaires. investigator explained the questionnaire and stayed in the room. The instruments used: the NO AZL Children's Quality of Life (TACQoL), the Trait Anxiety Inventory for children, and a specific FH survey. A random sample of 200 healthy Dutch children aged 12-16 served as a control for the TACQoL and Trait Anxiety Inventory surveys. Dirisamer, Simvastatin dosage begin at 5 mg/day for patients with LDL < 3 patients had abnormal levels of safety parameters: 2 patients (one 200370 220, and 10 mg/day for patients with LDL >220. If a range for receiving 5 mg/day and the other 10 mg/day) showed slightly higher LDL of 150-170 was not reached within the first 8 weeks, the values of CK. One (10 mg/day) expressed transiently elevated daily dosage was increased stepwise up to 20 mg/day. concentrations of ALAT (GPT) and GCT. Two boys (one 5 mg/day and the other 10 mg/day) suffered from Safety was assessed at follow-up visits every 4-8 weeks. Each headache in the morning of the first days treatment but which time patients were questioned about any adverse or unusual disappeared after a couple of days. signs or symptoms. Lab parameters include routine hematology One girl (5 mg/day) reported myalgia for about 2 weeks. and biochemistry, including blood urea nitrogen, creatinine, Two other patients (one male 10 mg/day, one female 5 mg/day) had electrolytes, total protein, albumin, fasting glucose, complete gastrointestinal complaints, which both resolved after 2 days without blood count, AST, ALT, alkaline phosphatase, CK, and total and changing the medication. direct bilirubin. There was no difference in mean BMI between baseline and 1 year. 7 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment De Jongh, 101/106 drug completed period 2002b69 1 and 83/106 completed period 2 (78%). 56/69 placebo(81%) completed period 2. De Jongh, Not reported. Methods of data collection of control 200370 subjects was not adequately described. Dirisamer, "Simvastatin was well tolerated. 200370 Side-effects were few and equally distributed among the three dosage periods." 8 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Ducobu, Simvastatin use in children Open-label To describe the long-term efficacy and safety data 24-36 months 1992243 (24-36 months) in 32 hypercholesterolemic children younger than 17 years. Hedman, Pharmacokinetics and Open-label clinical follow- To determine the single-dose pharmacokinetics and 8 weeks 2003240 pharmacodynamics of pravastatin up trial the lipid-lowering effect and safety of short-term use in children with familial Finland of pravastatin in children with heterozygous FH. hypercholesterolemia Knipscheer, Short-term efficacy and safety of double-blind, randomized To assess the safety, tolerability and efficacy of 12 weeks, following 199671 pravastatin in 72 children with placebo controlled study, pravastatin for HeFH an 8 week diet and familial hypercholesterolemia children stratified by age. placebo run-in 4 arms (3 doses of period. pravastatin, 1 placebo) Lipid clinic, The Netherlands 9 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Ducobu, Male or female children with total cholesterol >300 mg/dL after diet therapy for 32 Age <17 1992243 6 months. 22 (68.8%) male 10 (31.2%) female Hedman, 1. Age 4 or older 20 13 girls and 7 boys with 2003240 2. Total serum cholesterol > 6 mmol/L despite dietary interventions heterozygous FH. 3. Diagnosis of FH confirmed by LDL receptor mutation analysis or Age range: 4.9 - 15.6 years. lymphocyte test 4. Healthy, with no use of cholesterol lowering medication before study Knipscheer, Children with "known heterozygous FH" were eligible. HeFH defined as 72 children with HeFH 25 male; 47 female 199671 plasma LDL above 95th percentile for age and sex during lipid-lowering diet 66/72 White; 5/72 Black. AND hypercholesterolemia in siblings, parents or grandparents, or clinical Mean age 11.9-12.1 across manifestations of premature atherosclerosis prior to age 50 in 1st/2nd degree 4 groups. relatives. Exclusions: 1. Major surgery within the past 3 months 2. Use of medications interfering with lipid metabolism (anticonvulsants, oral contraceptives, corticosteroids, fibric acid derivatives or immunosuppressants). 3. Hepatic or renal dysfunction 10 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Ducobu, Children under age 10 started simvastatin at an initial dose of 5 Transaminases, alkaline phosphatase, and creatinine phosphokinase 1992243 mg/day. Titration to 10 mg daily after 4 weeks and to 20 mg after were measured at weeks 4, 12, 26, 52, 78, and 104, No significant another 4 weeks was possible. Older children began on 10 changes were detected. Transient increases in 1 transaminase, and 2 mg/day, increased to 20 mg/day after 6 weeks, and if necessary CPK measurements. to 40 mg/day after an additional 6 weeks. Growth: Children for whom height (n=12) and weight (n=16) were Mean dose: 16 mg/day. available remained in the growth percentages they had been in at baseline. Hedman, 10 mg pravastatin orally per day for 8 weeks. Few transient adverse events. No increase in serum alanine 2003240 aminotransferase, creatine kinase, or creatinine. Adverse events reported in questionnaires: abdominal pain (1), loose stools (1), headache (4), sleep disturbance (2), muscle tenderness or pain in rest (1), muscle tenderness or pain associated with physical training (1). All symptoms mild and disappeared during first 4 weeks of treatment. Knipscheer, Pravastatin 5, 10 and 20 mg/day Total number adverse events P- 9, 5 mg/day - 3, 10 mg/day - 6, 20 199671 mg/day - 1. List: rash, fatigue (P only), nose bleeding, headache, diarrhea (P only), dyspepsia (P only), nausea/vomiting, abdominal pain, myalgia (P only). Laboratory safety measurements did not show significant changes in any of the groups between the end of the treatment period and baseline. However CK abnormal in 8 of P , 6 of 5mg/day, 11 of 10 mg/day and 8 of 20 mg/day. Cortisol abnormal in 2 of P, 2 of 5 mg/d, 5 of 10 mg/d, 3 of 20 mg/d. 11 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Ducobu, Author states "well tolerated" This study was published only as a letter 1992243 Hedman, Author states "well-tolerated Duration only 8 weeks. Dose too low to 2003240 without any clinical significant achieve therapeutic response except for side effects" 2 patients - therefore, AE's associated with a therapeutic dose were not adequately assessed. Knipscheer, Compliance was 93% 199671 12 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Lambert, Treatment of familial RCT of drug To determine the efficacy, safety and tolerance of 8 weeks 199672 hypercholesterolemia in children multicenter: 6 sites in short-term administration of lovastatin in a male and adolescents: effect of Canada pediatric population with severe FH, and to evaluate lovastatin dose-response relationship. Boys aged 17 or younger with heterozygous FH McCrindle, Efficacy and safety of Atorvastatin RCT of drug To determine the safety and efficacy of atorvastatin 26 weeks 200376 in children and adolescents with 20 centers world-wide (US, (10 to 20 mg) in children and adolescents with familial hypercholesterolemia or Canada, Europe, South familial hypercholesterolemia (FH) or severe severe hyperlipidemia: a Africa) hypercholesterolemia multicenter, randomized, placebo- controlled trial Children aged 10 to 17 with FH or severe hypercholesterolemia 13 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Lambert, 1. Weight 27 kg or more 69 randomized, 10 mg/day group 199672 2. Plasma LDL above 95th percentile for age while on lipid-lowering diet 17 to 10 mg/day group Mean age: 12.5 (2.4) 3. History of unsuccessful treatment with bile acid-binding resins 18 to 20 mg/day group Female gender: 0 4. Family history of atherosclerosis at or before age 50 19 to 30 mg/day group Race: 94.1% white 5. Documented family history of hyperlipidemia with LDL > 95th percentile for 15 to 40 mg/day group 20 mg/day group age and sex before treatment or a personal diagnosis of FH substantiated by Mean age: 12.7 (1.6) LDL receptor activity or mutation. Female gender: 0 6. No significant concomitant conditions Race: 100% white 7. Weight and height not 3 - 97th percentile for age 30 mg/day group Mean age: 13.3 (2.7) Female gender: 0 Race: 94.7% white 40 mg/day group Mean age: 12.9 (2.7) Female gender: 0 Race: 93.3% white McCrindle, 1. Known FH or severe hypercholesterolemia and LDL-C>190mg/dL or LDL-C 187 Atorvastatin group 200376 > 160 mg/dL and a family history of FH or documented premature CV disease randomized, Mean age: 14.1 (2.0) in a 1st or 2nd degree relative; 2. TG < 400 mg/dL; 3. >Tanner stage II; 4. Not 140 to drug Female gender: 32% premenarche, pregnant or breastfeeding; 5. Testicular volume > 3cm3 after 47 to placebo Race: W: 94% age 12; 6. Weight between 10th and 95th percentile for age; 7. No liver or B: 1.4% kidney disease; 8. No known sensitivity to statins A: 1.4% Placebo group Mean age: 14.1 (2.2) Female gender: 28% Race: W: 87% B: 2.4% A: 2.4% 14 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Lambert, After pre-study visit and lab tests, all lipid-lowering medications No serious clinical adverse effects reported. Three patients had > 199672 were discontinued at least 8 weeks before study start, with no asymptomatic elevations in creatine kinase, which spontaneously lipid-lowering therapy allowed other than that permitted by study returned to normal, with no action required regarding the drug. protocol. Placebo given weeks -4 to 0, followed by randomization to double-blinded active treatment of lovastatin at 10, 20, 30 or 40 mg/day for 8 weeks. Medication dispensed at weeks -4, 0, and 4. Compliance verified by tablet count at weeks 0, 4 and 8. Clinic follow-up visits at weeks -4, -2, 0, 2, 4, 6, and 8. Complete physical exam at weeks 0 and 8. Ophthalmologic exam during placebo period and within 8 weeks of study exit. Patients continuously counseled by dietician to follow lipid-lowering diet throughout the trial. Daily food records completed by patients for 3 consecutive days and reviewed by a dietician at weeks -4, 0, 4, and 8. McCrindle, After a 4 week baseline/placebo phase in which subjects were No significant adverse effects documented. Incidence of treatment- 200376 instructed to follow the NCEP step 1 diet, those whose LDL-C related adverse events during double-blind: 7% in drug vs. 4% in remained > 160 mg/dL and had TG < 400 mg/dL at week -2 were placebo group (p=.7) Most adverse effects were mild or moderate. randomly assigned in 3:1 ratio to receive 26 weeks double blind Drug had no effect on sexual development. treatment with drug (10mg/d) or placebo. Drug could be titrated Labs: 1% on atorvastatin had elevated AST, and 1% had elevated to 20 mg/d at week 4 for those not at LDL-C < 130 mg/dL. Those ALT. completing double-blind were eligible to continue treatment for 26 weeks with open label atorvastatin (10 mg/d). 15 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Lambert, Drug well-tolerated. 85% of No separate control group: patients 199672 patients took at least 70% of the served as own controls prescribed dose during the active treatment period. McCrindle, Drug tolerated as well as 200376 placebo. 16 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Sinzinger, Professional athletes suffering Descriptive study in To monitor the experiences of 22 adolescent and 8 years 2004242 from familial hypercholesterolemia professional athletes with young-adult professional athletes with FH, in whom rarely tolerate statin treatment FH; treatment with statins was attempted. because of muscular problems. Austria Stefanutti, Diet only and diet plus simvastatin Non-randomized To investigate the effect of hypolipidemic treatment - 12 months 1999244 in the treatment of heterozygous controlled trial low cholesterol diet only (group A) or diet plus 10 familial hypercholesterolemia in mg/day simvastatin (Group B). childhood 17 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Sinzinger, Patients were considered as professional athletes when they had attended an 22 Mean age 24.1 (range 15 - 2004242 Austrian championship at any age class during the last 2 years or were playing 27) gin the top two leagues of their respective discipline. All were suffering from 31.8% female FH as diagnosed at the receptor level. Testing for anabolic steroids was done in all athletes to exclude any possible influence. Stefanutti, Pediatric patients with heterozygous FH 16 Age 7-12, mean 8.75 1999244 7 (44%) males 9 (56%) females 18 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Sinzinger, No drugs including vitamins were taken for at least 4 weeks. Among 87 periods of attempted statin therapy, the treatment was 2004242 The lowest available dose of statin was used as a starting dose. tolerated only 16% of the time. Muscle pain (weakness, cramps, Subjects were given atorvastatin, fluvastatin, lovastatin, aches, others) was reported in 84% of periods of statin therapy. The pravastatin, or simvastatin. Switching patients who tolerated a mean time of onset of muscle pain was 8.3 days (range 2 to 18 days). statin occurred only if they did not achieve target values. The shortest duration of treatment before switching was 8 weeks. Only two of the 22 subjects tolerated each of four statins (lovastatin, Safety was assessed by blood samples for CK and liver enzymes pravastatin, atorvastatin, and simvastatin. Two other patients tolerated (GOT, GPT, gamma-GT) drawn regularly atorvastatin and fluvastatin, but experienced muscle pain on lovastatin, pravastatin, and simvastatin. Three patients had elevated CK with one or more statins. An increase in liver enzymes was not observed in any of the athletes. Stefanutti, Patients were submitted to a 3-month washout (free diet). Then No adverse effects were observed during the therapeutic approach 1999244 they were given an AHA Step 2 diet for 6 months: cholesterol with diet, and with diet plus simvastatin, on relatively long-term <=200 mg/day, saturated fats <=10% of calories, total fats treatment. <=30% of total calories. After 6 months they were divided in two groups, matched for sex, age, and body mass index: Group A (n=8) received only the diet. Group B (n=8) was given 10 mg/day simvastatin. Followup lasted 1 year for both groups. 19 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Sinzinger, When initiating a statin therapy, 2004242 only 3 out of 22 athletes (11%) tolerated the chosen drug. Another 3 patients tolerated at least one statin. Only two athletes tolerated all the statins used. Stefanutti, Patients showed good Small sample size 1999244 compliance with the treatment. 20 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Stein, 199978 Efficacy and safety of Lovastatin RCT of drug To assess the lipid-lowering efficacy, biochemical 48 weeks in adolescent males with 14 pediatric clinics in the safety, and effect on growth and sexual development heterozygous familial USA and Finland of lovastatin in adolescent boys with HeFH hypercholesterolemia Wiegman, Efficacy and safety of statin RCT of drug To determine the 2-year efficacy and safety of 2 years 200437 therapy in children with familial Netherlands pravastatin therapy in children with familial hypercholesterolemia FH children ages 8 to 18 hypercholesterolemia (FH) recruited from academic medical referral setting 21 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Stein, 199978 1. Followed American Heart Association (AHA) pediatric diet for at least 4 132 randomized, Lovastatin group months 67 to drug Mean age: 13.3 (0.3) 2. LDL-C 189 mg/dL - 503 mg/dL 65 placebo Female gender: 0% 3. At least 1 parent had an LDL-C of at least 189 mg/dL not due to secondary Race: 93% of all were white causes or LDL-C values were 220 - 503 mg/dL or their LDL-C values were 220 - 503 mg/dL and a parent died of CAD with no available lipid values. Placebo group 4. No homozygous FH, underlying disorders known to produce secondary LDL- Mean age: 13.1 (0.3) C elevations Female gender: 0% 5. No disorders affecting triglyceride-rich lipoproteins Race: 93% of all were white 6. After the trial began, FDA requested only subjects who had Tanner stage 2 at entry be included in the trial. 7. No delayed puberty shown by testicular volume < 3 cm after age 12 8. Weight > 32 kg and between 10th - 95th percentile for age Wiegman, 1.Parent with clinical or molecular diagnosis of FH; 2.Two fasting samples with 214 randomized, 106 to Pravastatin group 200437 LDH-C levels of 155 mg/dL or more and triglyceride levels below 350 mg/dL; drug Mean age: 13.0 3.Adequate contraception for sexually active girls; 4. No drug treatment for FH 108 to placebo Female gender: 57% or use of plant sterols; 5. No homozygous FH; 6. No hypothyroidism; 7. No Race: NR abnormal levels of muscle or liver enzymes. Placebo group Mean age: 13.0 Female gender: 57% Race: NR 22 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Stein, 199978 All received formal dietary reinstruction, were monitored, and 1. 1 drug and 2 placebo subjects discontinued due to AE's. No determined to be stable following a AHA pediatric diet for at least clinically significant AE's. 8 weeks before randomization. Placebo was administered week - 6 withdrew in lovastatin, 1 due to AEs: increased bruising and purpura 4 to week 0. All subjects continued diet instruction, monitoring, 16 withrew in placebo, 2 due to AEs: skin rash and myalgia and evaluation throughout the study. 2. Drug had no significant effect on growth parameters at 24 or 48 Randomized to drug: Lovastatin, starting at 10 mg/day, with a weeks forced titration at 8 and 16 weeks to 20 and 40 mg/day, 3. No difference between drug and placebo in testicular volume, respectively. 40 mg/day weeks 25 - 48. Tanner staging, testosterone, or LH level. DHEAS increased 18% in Randomized to placebo: Matching placebo tablets the drug group compared to 5% in the placebo group. (p=.03) 4. Reduction of tocopherol in drug group at week 48 (p=.002), consistent with decrease in LDL-C. In both groups, increase in ALT from baseline, with no differences between groups at week 48 (p=.20), although increase was significant for each group (drug, P<.001; placebo, P=.008). No consistent changes in AST or CK. Report of infrequent, sporadic, nonsustained CK elevations in response to exercise. Wiegman, After initiation of a fat restricted diet for 3 months or more and No differences between groups in changes in height, weight, BMI, body 200437 encouragement of regular physical activity, randomized to surface area, testis volume, liver and muscle enzymes, and endocrine pravastatin 20 - 40 mg/day or placebo. Evaluated every 6 months function parameters (corticotropoin, cortisol, DHEA-S, FSH, LH, for 2 years. thyrotropin, 17b-estradiol, testosterone), and Tanner Stage. One child had an asymptomatic but extreme CPK elevation (16400 U/L) after 168 of placebo treatment. CPK level decreased to normal within 1 week of stopping treatment. No differences observed in growth, endocrine function parameters, Tanner staging scores, onset of menses or testicular volume. 3. No differences observed in muscle or liver enzyme levels. 23 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Stein, 199978 6 withdrew in lovastatin, 1 due to Good AEs: increased bruising and purpura 16 withrew in placebo, 2 due to AEs: skin rash and myalgia Wiegman, Tablet counting revealed that 2 year follow-up 200437 84% of tablets were taken; Well-designed mean visit attendance per child was 95% of all study visits. 24 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Bile-acid binding resins Curtis, 1991245 Loss of dental enamel in a patient Case report Report of a case of extensive loss of dental enamel 2 years of treatment taking cholestyramine in a young boy, thought to be attributed to mixing the with cholestyramine cholestyramine in Kool-Aid and swirling the mixture in the mouth before swallowing. Farah, 1977246 A study of the dose-effect Descriptive study 1) To determine the optimal dose of cholestyramine 21 days relationship of cholestyramine in Lipid clinic, Johns Hopkins required in children to significantly lower their LDL children with familial Hospital and cholesterol; 2) to relate this dose to their body hypercholesterolemia. weight and pre-treatment levels of LDL and cholesterol; 3) to study the effectiveness of a dose given twice daily; and 4) to monitor for potential side- effects, such as fat soluble vitamin malabsorption. 25 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Bile-acid binding resins Curtis, 1991245 N/A 1 7-year-old boy, LDL 240 mg/dL. Farah, 1977246 Children and young adults with HeFH were identified through screening 20 Mean age 15 y children of affected parents followed at the Lipid Research Clinic. 26 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Bile-acid binding resins Curtis, 1991245 The family was instructed to follow AHA Step 1 diet, and an After 2 years, the parents noted a change in the color of the patient's aerobic exercise program. After 6 months the patient's LDL was teeth, and a subsequent dental examination revealed extensive loss of 255 mg/dL, and so was given 4g of cholestyramine daily twice enamel. Serum calcium, phosphorus folate, and vitamin B12 values daily for the next 2 years. The parents stated that they had were normal. Results of a radioactive bone mineral analysis were mixed the cholestyramine in Kool-Aid to improve the taste, and normal. the patient stated that he had "swished" the mixture in his mouth KoolAid has a pH of 2.9 after water is added; the pH decreases to 2.4 for 10 to 15 minutes before swallowing. after cholestyramine is added. Farah, 1977246 13 patients were maintained on an outpatient isocaloric diet None of the patients on cholestyramine demonstrated any significant (<200 mg cholesterol with polyunsaturated: saturated fat ratio of side-effects such as fat-soluble vitamin malabsorption. 2.0). The diet was continued for 21 days. After 5 days, cholestyramine was started at 1 g/day and increased by 1 g/day up to a total of 16 g/day given twice daily. In another group, 7 children received the same dietary treatment, and started on a dose of cholestyramine predicted by a regression based on post-diet level of LDL: either 4 g, 8 g, or 16 g/day. 27 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Bile-acid binding resins Curtis, 1991245 NR Case report Farah, 1977246 NR Reports on different AEs in the same study as Farah, 1997 below. 28 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Farah, 1977247 Dose-effect relation of Descriptive study 1) To determine the minimum dose of 21 days: 5 days of (See Farah, cholestyramine in children and Lipid clinic, Johns Hopkins cholestyramine required to lower significantly the baseline evaluation 1977 above) young adults with familial Hospital plasma LDL cholesterol and total cholesterol in followed by a 16- hypercholesterolemia young patients with FH; 2) to relate this dose to their day period of drug body-weight and pretreatment concentrations of LDL treatment. cholesterol and total cholesterol; 3) to study the effectiveness of the dose given twice daily; 4) to monitor for immediate side-effects such as gastrointestinal symptoms and folate deficiency. Glueck, 1973263 Pediatric familial type II Noncomparative clinical To evaluate the effects of diet and diet plus 6 months with hyperlipoproteinemia: therapy study cholestyramine on cholesterol and LDL in 36 children additional 12-18 with diet and cholestyramine resin with familial type II hyperlipoproteinemia from 18 months follow-up for kindreds with documented familial type II. growth 29 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Farah, 1977247 Criteria for FH were the presence of xanthomas and 20 Mean age 15 ±5 S.D. years (See Farah, hyperbetalipoproteinaemia in the parent, or the transmission of 1977 above) hypercholesterolemia and hyperbetalipoproteinemia through many generations. The latter conditions were defined by concentrations of plasma total and LDL cholesterol above 95th percentiles, adjusted for age and sex. The pattern of inheritance of hyperlipidemia was not indicative of familial combined hyperlipidemia in any of these kindreds; none of the affected members had type-III hyperlipoproteinemia. Secondary causes of hyperlipoproteinemia were excluded in each patient by tests of thyroid, renal, and liver function. Glueck, 1973263 36 children aged 7-21 from 18 kindreds with well-documented familial type II 36, including 20 on diet Age 7-21 hyperlipoproteinemia were studied. These children had been originally found plus cholestyramine to b e heterozygous for familial type II hyperlipoproteinemia. Excluded secondary hypercholesterolemia. 30 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Farah, 1977247 Patients had been treated for a minimum of 6 weeks on a In one patient, febrile gastroenteritis developed after 7 days of (See Farah, therapeutic outpatient diet. NIH-type II diet was calculated daily treatment with cholestyramine, and resulted in therapy being stopped. 1977 above) for each of the children during admission to the metabolic unit. Mean serum-folate among 15 patients before treatment was 9.0 Average cholesterol intake was 159 mg/day (range 118-200) with ng/mL, and did not differ significantly between males (mean 9.8 ng/mL, mean ratio of polyunsaturated to saturated fats of 2.1 (range 1.9 N=5) and females (mean 8.6, N=10). Serum folate was measured to 2.2). Thirteen patients were given cholestyramine, with dose after treatment in 11 patients. Analysis by sex showed an upward increased by 1 g daily until a total dose of 16 g/day was trend for 4 males but a significant decrease in serum-folate in 7 achieved. Mean dose of cholestyramine = 11 g/day. females. SGOT increased in 2 patients (by 32 and 41 i.u./L). Two other patients had significant increases in LDH (>70 I.U./L). These results were not accompanied by any significant changes in other tests of hepatic function. The increases in SGOT persisted for 6 months then became normal. Increases in LDH were only transient. No other abnormal chemical measurements were observed. Glueck, 1973263 Children and their parents were instructed to ingest a low 10 of 12 routinely took the suggested dose of cholestyramine. cholesterol type II diet (<300 mg/day), and were seen monthly Weight gain and growth progressed during the 12 and 18 month follow- thereafter. At monthly visit, diet adherence was reviewed with a up period along normal gwoth percentile. No consistent changes in written 24-hr diet recall, separately with the children and their any of the laboratory safety tests (CBC, liver function tests, BUN, mothers. After 6 months of diet, the children were arbitrarily calcium, phosphorous, total prtein, electrolytes and carbon dioxide). separated into two groups, diet responders (n=11, LDL < 170 No evidence for hypercholoremic acidosis at any time. mg/dl) and diet non-responders (n=25, LDL 170+ mg/dl). 20 of the diet non-responders then received cholestyramine therapy (12 g/day) in addition to diet, and were seen monthly with review of adherence. Safety measures at monthly visits included height, weight, blood pressure, physical exam, drug and diet adherence. Lab tests were obtained at every other visit, including full hemogram, liver function tests, blood urea nitrogen, calcium, phosphorous, total protein (albumin/globulin), electrolytes, and carbon dioxide. 31 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Farah, 1977247 One subject stopped (See Farah, cholestyramine therapy due to 1977 above) febrile gastroenteritis. Glueck, 1973263 Of the 20 children on cholestyramine, 10 routinely took the suggested dose and also adhered to diet. 10 others had good diet adherencew but did not consistently take the full cholestyramine dose. 32 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Glueck, 1974262 Plasma vitamin A and E levels in Noncomparative clinical To evalute the short-term effectgs of diet and diet average follow-up of children with familial type II study plus cholestyramine on plasma vitamin A and E 6 months hyperlipoproteinemia during levels in 46 children with familial type II therapy with diet and hyperlipoproteinemia, and in 36 normal children. cholestyamine resin Glueck, 1977260 Therapy of familial Noncomparative clinical To determine whether a higher dose of 16 children followed hypercholesterolemia in study cholestyramine (16 mg/day) would provide a more for 18 months; 12 childhood: diet and substantial cholesterol level-lowering effect. for 24 months; 7 for cholestyramine resin for 24 to 36 30 -36 months months Glueck, 1986261 Safety and efficacy of long-term Observational study To examine the safety and efficacy of long-term diet 4.3 years diet and diet plus bile acid-binding and diet plus BABR therapy for pediatric familial resin cholesterol lowering therapy hypercholesterolemia in 73 children heterozygous for in 73 chlidren heterozygous for familial hypercholesterolemia familial hypercholesterolemia 33 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Glueck, 1974262 Diagnoses of hypercholesterolemia and hyperbetalipoproteinemia were made 46; of whom 30 were on Age 5-21 in 46 children, aged 5-21, from 19 kindreds with famlial type II diet plus BABR hyperlipoproteinemia. These children were heterozygous for familial type II and the primary nature of their hyperbetalipoproteinemia was confirmed by exclusion of common diosrders which produce secondary hypercholesterolemia. 36 normocholesterolemic siblings of the hypercholesterolemic children were also studied. Glueck, 1977260 Sixteen children heterozygous for familial hypercholesterolemia (aged 9-17) 16 (same children as in Age 9-17 from six kindreds with well-documented familial hypercholesterolemia were Glueck 1973 above, who studied. Their ascertainment, the nature of their hypercholesterolemia, their were on diet plus response to diet, and their short-term response to cholestyramine resin cholestyramine) (12g/day) have been previously described. (see Glueck, 1973) Glueck, 1986261 The 73 children were heterozygous for familial hypercholesterolemia as 73, including 33 on diet mean age 10.3 defined by primary elevations of total and/or low-density lipoprotein cholesterol plus resin therapy equal to or more than the age, sex, and race-specific 95th percentile, absence of secondary causes of hypercholesterolemia, and at least 2 similarly affected first degree relatives. 34 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Glueck, 1974262 Children were instructed to ingest a low cholesterol Type II diet. Plasma vitamins A and E remained within the normal range and Diets were not supplemented with adidtional vitamins A, D, K, or paralleled at higher levels the 25th to 75th percentile distirbution for E. Diet adherence was reviewed at monthly visits. The 46 vitamin A and E in normal children. hypercholesterolemi children were maintained on the diet for an average of 4 months per child, then divided into two groups: 16 whose TC & LDL normalized, and 30 whose TC and LDL did not normalize on diet. These 30 were then placed on cholestyramine in addition to the diet (12 g/day) and seen at monthly intervals, with average followup of 6 months on diet and cholestyramine. Glueck, 1977260 In this study, at month 13 of the 6-month Glueck 1973 study, the 1 child had persistent consitipation. None had nausea. 5 complained dose of cholestyramine was increased to 16 g/day and the that resin was gritty, had poor palatability. 1 complained of chronic children were followed up subsequently for an additional 1-2 fatigue. CBC, liver function tests, vitamin A and E, calcium, years. The children were seen in the outpatient research clinic phsophorus, blook urea nitrogen, fasting blood sugar levels did not every 4-6 weeks. At each visit, adherence was reviwed using 24- vary signficantly over time when compared to baseline levels on diet hour diet recall. Drug adherence was determined by packet only. count. Assessment of AES included clinical symptoms and questions about constipation, nausea, flatulence, overall palatability of the resin, and fatigue. 16 children were followed up for months 13-18, 12 for months 19-24, 7 for 25-30, and 31- 36. Glueck, 1986261 73 children heterozygous for familial hypercholesterolemia were Height and weight percentiles at completeion were no unchanged from compared with 39 normal healthy children wfrom a local pediatric baseline. No abnormal growth patterns noted. Sexual maturation was practice, matched for sex and age at entry, who were ingesting normal. None had unusual infections or extra school sick days. None habitual diets without modifications by family or physicians. of the girls had ammenorrhea; though 1 competitive cross-country Height and weight were obtained in both groups of children. runner had persistently irregular periods. No depression, suicide HeFH children were put on a modified type II diet. Dietary attempts or traumatic deaths. adherence was assessed at each visit. BABR therapy (cholestyramine or colestipol, 8 to 20 g/day) was not started until the children had been on diets alone for at least 4 months. No hypocaloric regimens were recommended, and supplemental fat-soluble vitamins were not given. Serum folic acid levels were not systematically measured. 35 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Glueck, 1974262 NR Glueck, 1977260 11 of 16 had good adherence. 5 children dropped out after 2 years because of poor palatability of the drug. Glueck, 1986261 4 children exhibited transient refusal to follow diets and/or take bil acid-binding resins. 36 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Groot, 1983248 The effects of colestipol Non-randomized, cross- To evaluate the effects of colestipol vs. placebo on 16 weeks hydrochloride on serum over placebo controlled. serum lipoprotein lipid concentrations in lipoprotein lipid and apolipoprotein in children ages 7-15 with heterozygous hypercholesterolemic patients B and A-1 concentrations in heterozygous familial consuming a type II-A diet children heterozygous for familial hypercholesterolemia type hypercholesterolemia II-A The Netherlands Hansen, Growth during treatment of Descriptive study To observe somatic growth in children treated for 8.5 years in 13 1992249 familial hypercholesterolemia Pediatric lipid clinic, FH. (diet); 5.5 years in Denmark 17 patients (diet followed by diet + colestipol) 37 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Groot, 1983248 Inclusion criteria: 33 children Not given 1. One parent with hypercholesterolemia or who had died from cardiovascular Two comparable groups disease. were formed with respect 2. Serum cholesterol levels >6.2 mmol/l and serum triglycerides normal (<1.5 to age, sex and serum mmol/L) after a low fat breakfast. cholesterol concentration. 3. No disease associated with secondary hyperlipidemia; no hypertension. One group given placebo then colestipol (8 weeks of each); the other group given colestipol then placebo. Hansen, Children with FH were followed in a pediatric lipid clinic for at least one year 30 Age range 1/17, median 4.5 1992249 from 1970-1987. Several children were ascertained through a neonatal pilot 53% female screening program. The diagnosis of FH was based on increased concentrations of serum total and low-density lipoprotein cholesterol in the child, in one of the parents and in other members of a pedigree compatible with autosomal dominant inheritance of type IIa hyperlipidemia. Secondary hyperlipidemia was excluded by obtaining normal results on laboratory tests of thyroid, renal, hepatic, and glucose metabolism. 38 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Groot, 1983248 Placebo was silicagel stained with FeCl3. Dose of colestripol or NR placebo was 5 g at AM and 5 gm at dinner for children <40 kg; and 10 gm at breakfast and 5 gm at dinner for children >40kg. Mean dose was 0.34+0.08 (mean+sd) g/kg/body weight/day. Hansen, Dietary counseling was provided by the physician in the lipid Among all 30 children, the SD scores for both height/age and 1992249 clinic. 17 children additionally received colestipol at 125-250 weight/age decreased by approximately 0.4 during dietary treatment mg/kg/day in 1-2 daily doses. (p<0.05), but were not affected by treatment with colestipol. The addition of colestipol to the dietary regimen of 17 children did not cause any statistically significant change of growth. No child decreased to below -2 SD during dietary treatment. During treatment with colestipol the height/age, but not the weight/age, of only one child decreased below -2 SD. A deficit of 0.4 SD score for a 10-year-old boy corresponds to approximately 1.8 kg in weight and 2.4 cm in height. 39 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Groot, 1983248 5 children withdrew because they didn't like the taste of the medication: 2 in the group that started with colestipol and 3 in the group that started with placebo. Hansen, NR 1992249 40 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Harvengt, Colestipol in familial type II Non-randomized, single- To evaluate the effects of colestipol on serum Up to 36 months of 1976250 hyperlipoproteinemia: a three-year blind trial with placebo lead cholesterol in patients with primary type II treatment, following trial in phase hyperlipoproteinemia. 6-wk placebo 13 patients aged 6 to 61 with familial type II hyperlipoproteinemia Koletzko, Treatment of Descriptive study, lipid To study the effects of outpatient treatment with diet Diet: 4-70 months 1992251 hypercholesterolemia in children clinic, Germany alone or with combined diet and drug therapy in (mean 17.5 months) and adolescents pediatric outpatients. With genetic hypercholesterolemia. Combined therapy, mean observation period was 27.9 months 41 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Harvengt, Inclusion criteria: Included 3 children: ages 3 pediatric patients: 1976250 Hypercholesterolemia or tendon xanthomas or both. Type II lipoproteinemia 6, 11, and 18 1 male aged 6 was defined by serum cholesterol and triglyceride levels, agarose lipoprotein 2 females age, 11 and 18 electrophoresis pattern, clinical signs (corneal arcus, xanthomas), and familial study. Koletzko, Diagnosis of primary genetic hypercholesterolemia was based on a positive 35 on diet only Mean age 7.9 (range 2-17.6) 1992251 family history suggestive of dominant inheritance and on at least two fasting 14 on diet + BABR blood tests with results for total cholesterol, LDL cholesterol, and in the case of familial combined hyperlipidemia, triglycerides also, exceeding the 95th percentile for age and sex as well as a positive result of lipoprotein electrophoresis. 42 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Harvengt, All patients followed a diet less than 300 mg cholesterol daily There were only 3 children in this study, and no AEs were observed in 1976250 with polyunsaturated-saturated fatty acid ratio >1,8. Patients them except for low iron level without anemia was seen in an 18-yr-old were given an inert cellulose powder as placebo for 6 weeks. female. No hepatic dysfunction, renal function, or hematologic side Colestipol was then given before meals in 5 gm t.i.d. effects were observed. The prothrombin time values were not abnormally prolonged during colestipol medication. Fasting blood sugar increased from 74 mg/dl (placebo) to 88.5 mg/dl after 18 months of treatment, but returned to baseline values after 3 years. Serum uric acid level increased during treatment but did not reach abnormal values. There was no steatorrhea in patients receiving colestipol. Serum vitamin B12 and folic acid content were not impaired after 2 and 3 years on colestipol. Serum iron concentration was low in one pediatric patient, without hematologic signs of anemia. "There was some increase in weight in the two children." Mild gastrointestinal complaints (flatulence, constipation) were described during the first 3 months of treatment, which disappeared despite continuation of treatment. Koletzko, Diet focused on limiting intake of saturated fatty acids to <10% of No serious side effects were noted with both forms of treatment (diet/ 1992251 energy and replacing them with poly- and monounsaturated. Fats diet + BABR). "Mild gastrointestinal symptoms occurred in very few as well as reducing cholesterol intake to <100 mg/1000kcal. 14 patients." Percentile values for weight and length remained patients, whose serum TC and LDL remained severely increased unchanged in all patients with the exception of three overweight despite diet, additionally received cholestyramine 2-4 times daily children whose weight fell by more than 5 percentage points but (mean daily dose 0.36 g/kg) in addition to continued diet therapy. remained above the 50th percentile. 43 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Harvengt, NR 1976250 Koletzko, NR 1992251 44 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Liacouras, Use of cholestyramine in the Observational To report the experience with cholestyramine in Varied, 0-62 months 1993252 treatment of children with familial Philadelphia children over a 10 year period. combined hyperlipidemia Lipid clinic McCrindle, Acceptability and compliance with Randomized crossover To compare acceptability, compliance and 28 weeks 199774 two forms of cholestyramine in the Patients recruited from 2 effectiveness of two forms of cholestyramine resin in treatment of hypercholesterolemia lipid disorder clinics treatment of hypercholesterolemia in children. in children: A randomized, crossover trial 45 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Liacouras, Referred to Lipid Heart Clinic at Children's Hospital of Philadelphia (87 of 673 87 85 met criteria for HeFH 1993252 children referred met criteria) (n=36) or FCHL (n=51). 2 Heterozygous FH (HeFH) was diagnosed if the plasma LDL-cholesterol level others were included was above the 98th percentile in both the child and an affected parent and if because of family history; total cholesterol levels were greater than 300 mg/dL in another 1st degree thus 87 children were family member. treated with cholestyramine. Familial combined hyperlipidemia (FCHL) was diagnosed when both the child and an affected parent had a plasma LDL-cholesterol or triglyceride level FCHL group: Average age above the 90th percentile and a first degree relative had documented 10.4+3.9 years; male/female hyperlipidemia and at least one affected family member had a plasma ratio 18/19 triglyceride value >90th percentile. Exceptions: 2 children were included in the study because of a strong family HeFH group: average age history of premature heart disease. 10.7+4.3 years; male/female ratio 13/12. McCrindle, 1. Age 10 - 18 years 40 43% female 199774 2. 1 or more parent with heterozygous familial hypercholesterolemia type IIA or 2 siblings from 6 families IIB 3 siblings from 1 family 3. Fasting serum LDL of 130 mg/dl or more while on American Heart Median age at enrollment: Association step 2 diet 13 (10-18) 3. No contraindications to use of cholestyramine Median age at diagnosis: 9.5 4. No lipid-lowering agents other than those in study protocol (0.1-17.2) 46 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Liacouras, All children had failed diet alone. Of the 62 who continued more than 1 month on the drug, 15 had an 1993252 All continued to adhere to a Step 1 NCEP diet. adverse effect: Cholestyramine given as Questran, initially as 4 gm/day in 2 12 nausea, 2 abdominal bloating, 1 severe constipation. divided doses; then increased in increments of 4 gm/day every 8- 73% complained of poor palatability 12 weeks up to 16 gm/day unless LDL levels dropped to <130 No instances of elevated prothrombin times. mg/dL or side effects occurred. In four patients the dose was increased to greater than 16 gm/day. All patients were given a multivitamin at hospital. McCrindle, 6 week washout period followed by random assignment in pairs 38/40 (95%) completed both medication periods 199774 to either 8 gm/day of cholestyramine powder (4 gm packets) or Authors noted minor gastrointestinal complaints were frequent but did pills (1 gm tablets) for an 8 week period. Drug dispensed with not result in any drop-out random over count, and subjects instructed to return unused drug at end of drug period. Subject completed log book during the 1st week on amount of drug taken, acceptability and adverse effects, with questionnaire at weeks 4 and 8. After the first 8 week course of medication and assessment, there was another 6 week wash-out, followed by crossover to the alternate form of medication and assessment. 47 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Liacouras, 73% complained of poor Unclear duration of treatment; post 1993252 palatability treatment values represent lipoprotein profiles taken at the time of maximum After cholestyramine therapy decrease of LDL (6-12 months). was initiated 7 pts (8%) failed to No intention-to-treat analysis return for follow-up visits, 18 (21%) began medication but discontinued it within 1 months, 62 (71%) continued to take medication >1 months and complete f/u visits McCrindle, At end of study, 82% preferred No control group 199774 pill, 16% powder, and 2% Small sample size neither form. Mean (+/-SD) At baseline, pill group had perception of compliance as assessed by increased importance of diet amount of medication taken was greater for pills (61% +/-31% than powder (50% +/- 30%, p=0.01). Form of medication increased compliance by at least 25% for 16 patients (42%), 13 in favor of pills and 3 of powder. 48 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study McCrindle, A Randomized Crossover Trial of Randomized cross-over To determine the short-term safety and effectiveness 18 weeks each 200275 Combination Pharmacologic trial of combination drug therapy in children period Therapy in Children with Familial Hyperlipidemia. Lipid clinic, Toronto, Canada Schwarz, Fat-soluble vitamin concentrations Pre/post To determine whether long-term administration of Up to 24 months 1980253 in hypercholesterolemic children Longitudinal colestipol hydrochloride to hypercholesterolemic treated with colestipol children along with a diet low in cholesterol is Outpatient clinic of effective in reducing plasma total and LDL, and to Washington University learn if changes in plasma concentrations of vitamins Lipid Research Center A, D, E, K or folic acid occur during therapy. Also from 1973 to 1978. assessed effect of drug on calcium metabolism by sequential determination of serum total and ionized calcium, phosphorus, alkaline phosphatase, and parathyroid hormone. 49 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics McCrindle, 1. Children between ages 8-18 40 randomized to either 11 girls, 25 boys 200275 2. Positive family history of hypercholesterolemia or premature atherosclerotic colestipol 10g/day alone median age 14, range 9-18 cardiovascular disease in first-degree relatives or colestipol 5 g/day BMI 22-25 kg/m2 3. Minimum fasting LDL cholesterol level before enrollment >4.15 mM/L +pravastatin 10 mg/day 4. Participation and compliance in dietary counseling program for at least 6 months. 5. No secondary cause for hyperlipidemia, no major surgery or serious illness within past 3 months. Schwarz, 1. Hypercholesterolemic (values > 95th percentile for age) children aged 5 - 17 23 Hypercholesterolemic 1980253 2. Plasma triglycerides normal children aged 5 to 17 years 3. One or more 1st degree relative with type II hyperlipoproteinemia (mean 12.2) (1exception) and 4 sibling controls 4. Secondary causes of elevated plasma concentrations eliminated 35% female 5. Cholesterol-lowering medications discontinued at least 3 months prior to study start 6. No supplemental vitamins or cholesterol drugs during study (except study drug) 50 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects McCrindle, colestipol 10g/day (10 tablets) alone vs. colestipol 5 g/day (5 Symptoms reported at the end of the study in the final preference 200275 tablets) plus pravastatin 10 mg/day (1 tablet) for first 18 week questionnaire showed that the majority of patients had no symptoms. period, followed by 8 week wash-out period and then each group 18% reported constipation with colestipol vs. 0% with the combination crossed over to the other treatment for the second 18 week vs. 3% who had it with both, 12% reported bloating/gas with colestipol period. alone vs. 0% with combination (3% had it with both). 21% had stomach ache with colestipol alone vs. 0% with combination vs. 0% with both. 11% had headache with colestipol vs. 0% with combination vs. 3% with both regimens. 6% had muscle aches on colestipol along vs. 3% with combination vs. 0% with both regimens. Schwarz, Child and parents instructed in NIH type IIA diet, and adherence No severe side effects reported. 1980253 was reviewed at each visit by 24-hour dietary recall. Subjects had 2 - 8 months (average 3.6) of diet only therapy, after which those 1 patient developed Reynauld's phenomenon on 4 separate occasions with cholesterol concentrations > 90th percentile for age were during therapy but continued to take drug without recurrence of given colestipol 15 - 20 gm/day. Drug therapy up to 24 months in symptoms. duration. Office visits every 2 months for diet and drug adherence checks and determination of total cholesterol and triglyceride levels in plasma. Blood work and physical exam performed twice during diet only period and at 6 monthly intervals during colestipol therapy. Measures: 1. 24 hour dietary recall 2. Package count of unused drug on > 3 separate visits 3. Detailed blood work: blood count, serum gluclose, urea nitrogen, glutaminic oxaloacetic transaminase, bilirubin, total and ionized calcium, phosphatase, parathormone, vitamin A, 25- hydroxycholecalciferol, prothrombin time, vitamin E, folic acid, and plasma lipid and lipoprotein quantification. 4. Complete urinalysis 5. Vitamin E analyzed by colorimetric assay in 1st part of study and microfluorometric assay in 2nd part. 6. Folic acid analyzed by Lactobacillus casei method in 1st part of study and later by radioimmunoassay. (Values converted) 51 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment McCrindle, Compliance was 57-66% overall 200275 and was similar between treatment groups and dose groups. There was a decrease in compliance between the first 8-week and second (10-week) dispensing periods, this difference was not significant. Schwarz, 6 subjects complained of poor Many uncontrolled variables: 10 patients 1980253 palatability of drug, and 4 had been on low cholesterol diets 1 - 4 withdrew for this reason. years prior to study start. Range of "diet alone" treatment was 2 to 8 months, with Only 7 patients completed 18 - an average of 3.6 months. Drug 24 months of drug (30%); 3 compliance in the good adherence group controls completed this (n=13) was 82% (range 100% - 63%), treatment segment and was undetermined in the poor adherence group. Poor adherence group attended 66% of visits. Only 7 patients (30%) completed 18-24 months of drug. Both folic acid and vitamin E were analyzed by different lab methods in different parts of the study. Results section confusing. Design included 4 sibling controls with no resulting data analysis on this group. No separate analysis or adjustment for the 1 subject who did not have a 1st degree relative with hyperlipoproteinemia. 52 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Tonstad, Colestipol tablets in adolescents Observational To examine the palatability and side effects of 6 months for 1996254 with familial hypercholesterolemia Oslo, Norway colestipol tablets in adolescents with FH and report colestipol; 2-10 changes in lipid levels (mean 6) years for diet Tonstad, "Low dose colestipol in DB RCT of colestipol 10g Effectiveness of colestipol in treating FH. 52 weeks 199680 adolescents with familial qd or 5g bid vs placebo hypercholesterolemia" 67 adolescents Norway 53 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Tonstad, heFH attending a referral lipid clinic; none were siblings. 27 children 23 boys and 4 girls ages 10- 1996254 16 Tonstad, adolescents age 10-16 referred to lipid clinic; all had two or more TC 74 targeted, 66 agreed to 37 boys (58%); 29 girls 199680 >6.2mmol/l and TG<3.0 mmol/l. participate mean age 13.2 Tanner stage 2.6 54 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Tonstad, All patients had previously been instructed to follow a diet All showed normal growth and development according to standard 1996254 restricted in saturated fat and cholesterol (<10% of caloric intake height, weight and pubertal staging charts over the 2-10 years that they and <200 mg/day respectively). had been followed prior to this study (on diet alone). All had previously taken colestipol granules or cholestyramine or Two girls taking 4 tablets daily had difficulty swallowing the tablets. both; all but 4 had stopped taking granules, usually because of One boy taking 6 tablets had abdominal discomfort at bedtime 2-3 unpalatability. times/week - he was taken off the med. 1 subject had difficulty with flatulence and swallowing the tablets. Colestipol was given as 2-12 g daily usually 6 g. Doses higher 24/27 reported that they forgot tablets 1-2 times/week or less. than 6gm/day were divided. Tonstad, All subjects instructed to follow a diet: 30% calories from fat; Total of 7 subjects dropped out 4 in colestipol; 3 in placebo. 5 pts 199680 <10% from saturated fat and <200 mg cholesterol/day. dropped out because they refused to swallow the drug; 2 dropped out Assessed with 4 day weighed food record. because they had unexpectedly low lipid concentrations at Subjects randomized to drug treatment got either colestipol 10g randomization. qd or 5g bid x 8 weeks. 8 girls did not agree to continue in the open phase. Open label trial after 8weeks in which placebo group started Of 51 in open phase, 42 completed 1 year. colestipol for 52 weeks; colestipol group continued on colestipol. 8 subjects in colestipol group had GI side effects attributed to the drug: 2 had constipation that improved with temporary dose reduction; 1 had dyspepsia and 1 had flatulence throughout entire study; 2 had intermittent nausea, 1 had temporary decrease in appetite; 1 had abdominal pain improved with dose reduction. Linear growth velocity was within reference range for Norwegian adolescents. 1 pt lost >1 kg during the study (24.5 kg/m2 pre and 21.3 kg/m2 post). Lab values: Low-dose colestipol (10 g/day) reduced concentrations of serum folate after 8 weeks. Serum vitamin E and carotenoids decreased proportionally with decreases in cholesterol. Vitamin D did not change significantly, but decreased more in subjects who were more compliant after 1 year. 55 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Tonstad, All preferred the tablets to 1996254 granules formulations of resins. 7 people did not complete 6 month labs (or at least are not reported). In SE table, 4 people noted to have stopped med prior to 6 months; reasons include unpalatable in all cases. Tonstad, Compliance was 68% in placebo group was older and more 199680 colestipol group vs 76% in advanced tanner stage (p<0.05) placebo group NS. Only one third of adolescents adhered to the drug regimen in the course of one year. Obstacles included vacations, special occasions, and unpalatability. 56 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Tonstad, Efficacy and safety of Random assignment to To determine the efficacy and safety of 1 year 199681 cholestyramine therapy in double-blind comparison cholestyramine therapy in young children with familial peripubertal and prepubertal Norway hypercholesterolemia children with familial hypercholesterolemia Boys aged 6 to 11 years and girls aged 6 to 10 years with familial hypercholesterolemia in a referral lipid clinic Tonstad, The C677T mutation in the Placebo-controlled RCT; To investigate whether the elevation of plasma total 1 year 1998255 methylenetetrahydrofolate lipid clinic, Norway homocysteine observed during cholestyramine (See Tonstad reductase gene predisposes to treatment was related to the 199681) hyperhomocysteinemia in children methylenetetrahydrofolate reductase genotype in with familial hypercholesterolemia children with FH. Cholestyramine may increase tHcy treated with cholestyramine by inhibiting the absorption of folate. West, 1973256 Use of cholestyramine in the Descriptive study To report the experience in the treatment of 19 Up to 20 months treatment of children with familial children with HeFH using cholestyramine. hypercholesterolemia 57 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Tonstad, 1. Familial hypercholesterolemia diagnosed if TC > 260 mg/dl and triglyceride 96 randomized, Drug group 199681 levels < 200 dl, if 1 parent had baseline cholesterol > 300 mg/dl and 36 to drug Mean age: NR triglyceride levels < 200 mg/dl, or tendon xanthoma, or if autosomal dominant 36 to placebo Female gender: 44% inheritance present in other members. Race: NR 2. Prepubertal girls aged 6 - 10, and prepubertal boys aged 6 - 11 who completed an initial medical evaluation and dietary session and had a parent Placebo group who understood the diet. Mean age: NR 4. Height velocity score standard deviation score > -2.0. Female gender: 33% 3. Those taking lipid-lowering drugs were required to discontinue doing so at Race: NR least 4 weeks before study start 4. Secondary hyperlipidemia ruled out by clinical and lab examination. Tonstad, Subjects were boys and girls aged 6 to 11 with FH who had participated in a 1- 96 Mean age 7.98 1998255 year open study of Step 1 diet. 40% female (See Tonstad 199681) West, 1973256 Most children were referred because of a family history of early onset of 19 children from 11 38.7% female ischemic heart disease. All were asymptomatic but one 13-year-old boy was families detected because of skin xanthomata behind the knees, and one 9-year-old girl had corneal arcus. All patients had characteristic biochemical evidence of familial hypercholesterolemia (i.e. raised TC and normal fasting TG, with increased B-lipoprotein by electrophoresis and density gradient ultracentrifugation). Also, at least one of the parents had the same lipoprotein abnormality or had died of ischemic heart disease. 58 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Tonstad, After 1 year of a low-fat, low-cholesterol diet, children with LDL > In the drug group, folate deficiency occurred in one subject, in whom 199681 190 mg/dl or > 160 mg/dl in the presence of familial premature the total homocysteine level increased concurrently. With the exception cardiovascular disease were randomly assigned to a double- of modest decreases in levels of vitamin D during the winter, other blind comparison of 8 gm cholestyramine or placebo for 1 year. nutritional deficiencies were not observed. Dieticians reinforced the diet at each visit. Dietary composition No adverse effects on growth and bone age evident. No difference in assessed by a 7 day record, evaluated by FIBER. Assigned dose mean height velocity standard deviation scores in those who hadn't 8 gm/day, following 1 week build-up phase of 4 gm/day. Follow- started puberty. There was one case of intestinal obstruction caused up visits scheduled after 2, 5, 7, 9 and 12 months. Compliance by adhesions assessed by counting leftover packages at each visit. One on cholestyramine vomited after taking 2 packets, and withdrew. One on placebo reported vomiting for 3 weeks and withdrew. Another on cholestyramine reported more frequent headaches and withdrew. Tonstad, Children were randomized to 8 g/day of cholestyramine or During cholestyramine treatment, tHcy concentrated increased in 1998255 matching placebo granules for 1 year. subjects with the C677T mutation in one or both alleles but not in (See Tonstad Blood was collected at baseline and 1-year follow-up, and subjects with the CC genotype. In contrast, there was a reduction in 199681) analyzed for plasma tHcy, serum folate, and MTFHR status. serum folate in most subjects, regardless of genotype. Serum folate was measured by radio assay. In the placebo group, tHcy and folate showed no consistent changes. West, 1973256 Cholestyramine twice daily in a total dosage of 8 to 24 g/day (0.3 Absorption of fat was impaired in some patients but has not been to 1.1 g/kg body weight per day). In 11 patients cholestyramine associated with diarrhea. Fecal fat was estimated in 9 patients therapy was combined with a diet low in saturated fat (<20 g receiving cholestyramine. 5 of 7 children on a normal fat intake had daily) and containing some supplementary corn oil and corn oil steatorrhea with mean fecal fat ranging from 5.3 to 10.5 g/day. Serum products. 8 patients remained on a normal diet during folate levels decreased in all patients, and 6 of 12 tested had cholestyramine therapy. In 4 patients a comparison of twice- subnormal red blood cell folate. There has been no evidence of daily therapy with 4-times daily was made, starting with a dose of malabsorption of other vitamins or of minerals. Growth rate has been 4 g 4 times daily before food, followed by 8 g twice daily (total normal in all children. dose unaltered). In all patients, growth and height and weight continued along the pretreatment centiles, with no change in velocity. None developed anemia, and hemoglobin, prothrombin time, serum calcium, alkaline phosphatase, and serum vitamin A concentrations have remained unchanged. 59 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Tonstad, 22 of 36 completed 1 year of 1. High attrition: only 22/36 (61%) in 199681 cholestyramine, having taken drug group and 26/36 (72%) in placebo 77% of all doses, and 26 of 36 completed the study. on placebo completed 1 year, 2. Only about half the children were able having taken 83% of all doses to take cholestyramine in the amount (ns). prescribed, primarily because of 12 of 14 withdrew on unpalatability. cholestyramine due to unpalatability within 2 months, as did 9 of 10 on placebo. Tonstad, 22 (61%) of 36 in the 1998255 cholestyramine group, and 26 (See Tonstad (72%) of 36 in the placebo group 199681) completed 1 year. Dropouts were usually due to unpalatability of the drug. West, 1973256 NR 60 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study West, 1975257 The effect of cholestyramine on Descriptive study To monitor the effect of long-term treatment on the 1 to 2.5 years intestinal absorption UK absorption of fat, vitamins, and some minerals among children with FH treated with cholestyramine. West, 1975258 Treatment of children with familial Observational To report the results of three different treatment Varied, 2-8 years hypercholesterolemia London regimens used to control cholesterol in children with heterozygous familial hypercholesterolemia (familial type II hyperlipoproteinemia) over an 8 year period. 61 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics West, 1975257 Children with the HeFH. IN the majority there was a history of premature 18 Age 1-14 years ischemic heart disease in a near relative. None of the children had any evidence of heart disease, two had corneal arcus, and one had skin xanthomata and was also overweight. West, 1975258 45 children with diagnosis made as the result of a family study of serum 45 children 23 boys, 22 girls ages 1-16 lipoproteins carried out because of premature coronary heart disease in a near relative (often a parent); diagnosis confirmed by identification of the lipoprotein abnormality in a t least one other first degree relative. 62 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects West, 1975257 Subjects were treated with Questran brand cholestyramine. For With prolonged treatment, folate deficiency occurred: mean serum 16 children, treatment was given twice daily in a dose of 0.2 to folate decreased from 7.7 ng/ml pre-treatment to 4.4 ng/mL for 1.1 (mean 0.6) g/kg/day. Two children took an equivalent dose 3 patients on treatment for over 1 year; a corresponding lowering of red times daily. Eight children remained on their normal diet, and 10 cell folate also occurred. Oral folic acid 5 mg daily overcame this were initially given a diet low in saturated fats supplemented with depletion. Prothrombin time remained normal in all patients. There corn oil and corn oil products. No patient received vitamin or was a significant decrease in mean serum levels of vitamins A and E mineral supplements. For most patients, hemoglobin, and of inorganic phosphorus over the first 2 years of treatment, prothrombin time, plasma calcium, phosphorus, protein, and although values remain within the normal range. In children on a alkaline phosphatase, serum folate, iron, and vitamins B12, A, normal intake of dietary fat, five of 7 tested had fecal fat >5g/day while and E were estimated in the majority before treatment and at 3 on cholestyramine. No child developed diarrhea, and growth has been month intervals thereafter. Red blood cell folate was not normal. Levels of serum iron, vitamin B12, plasma calcium, and determined at baseline was later added to the investigations. protein did not change significantly. West, 1975258 Three different regimens: Diet: lack of compliance 1. Diet modification alone (17 children): Diet consisted of a Clofibrate: intolerance of drug reduction in ordinary fat intake to about 18 g/day, and use of corn No obvious short-term side effects were apparent with diet alone, or oil. diet and clofibrate, but adherence to these regimens was poor. The 2. Dietary modification with clofibrate (9 children): clofibrate was growth rates of the 3 children maintained on diet alone have been administered 18-28 mg/kg/day; diet same as that above. All of normal. these children failed diet alone before starting clofibrate. Cholestyramine: unpalatable in some. Side effects have included 3. Cholestyramine (36 children): given as Questran in a total folate deficiency, steatorrhoea, and reduction in serum levels of daily dose of 0.3-1.1 g/kg/day given in two equal divide doses in vitamins A and E and of inorganic phosphorus although not to most children. Initially, low saturated fat diet was given with abnormally low values. cholestyramine but then changed to allow most children to (see withdrawals) consume a reasonably normal diet. Cholestyramine: Several children were managed sequentially be more than one regime. 63 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment West, 1975257 NR West, 1975258 Diet: By the end of 18 months, Poor - can't tell really who got what, for only 20% of the group were how long, or which children switched either satisfactorily maintained regimens. (at a cholesterol 10% lower than baseline). Clofibrate: By 2-1/2 years, no child was still on the drug (stopped on their own, or stopped because of inadequate cholesterol lowering). Cholestyramine: A few children stopped the drug after a short period because they found in unpalatable. After 2 years of treatment 72% remained satisfactorily controlled and after 3 years this number was 57%. 64 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study West, 1980259 Long-term follow-up of children Observational To assess the impact of cholestyramine in children 1-8 years with familial hypercholesterolemia London with FH. treated with cholestyramine Other drug therapies Baker, 198254 Treatment of homozygous familial Open-label clinical study To evaluate the effects of probucol in 10 HoFH 15-21 months hypercholesterolemia with patients probucol 10 HoFH patients aged 8- 46, of whom 5 underwent portacaval shunt operation prior to probucol treatment. 65 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics West, 1980259 children with FH as defined by 1) elevated plasma beta lipoprotein and 35 children 16 girls, 19 boys cholesterol concentrations (cholesterol above 7 mmol/L or 270.7 mg/dL) in the ages 1.3-17.4 years (mean patient, in the absence of other disease that could cause hyperlipidemia and 8.4) with a first-degree relative who either had the same pioprotein disorder or who 8 sibling pairs. had died of IHD before age 45. Other drug therapies Baker, 198254 Homozygous familial hypercholesterolemia based on presence of i) serum TC Included 7 7 pediatric subjects: consistently > 14.3 mmol/l, ii) appearance of xanthomas in the first decade of children/adolescents 2 males aged 8 and 16 life, and documentation in both parents of hypercholesterolemia or clinical aged 6, 8, 9, 11, 16, 18, 5 females aged 6, 9, 11, 18, signs indicative of the heterozygous state. 21 . and 21 66 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects West, 1980259 For 8 children cholestyramine was started as initial therapy. Almost all children expressed some dislike of cholestryamine, and a For 27 patients initial treatment had been with a low saturated fat few complained of transient gastric fullness after taking a dose. No diet-- this either could not be tolerated long term or failed to lower patient had persistent constipation, loss of appetite or diarrhea. No the cholesterol levels below 6.5 mmol/L (251.3 mg/dL). other symptoms were reported. Cholestyramine was started at 2 g or 4 g twice daily before One girl (age 18.6 yrs) stopped medication after 79 months on meals, the dose gradually increased until either the plasma- cholestryamine because she developed nausea, dizziness and cholesterol concentration was below 250 mg/dL or until the malaise. maximum tolerated dose was being taken. All patients were given 5 mg folic acid supplements daily. Other drug therapies Baker, 198254 All patients were kept on a low-cholesterol, low-saturated-fat diet One patient developed nausea on 500 mg bid after 10 months, and the supplemented with 30-60 ml maize oil daily for at least 3 months dose was reduced to 375 mg bid for 5 months; thereafter was given before starting probucol treatment. Two patients aged 8 and 9 500 mg/day without recurrence of nausea. No significant changes were given probucol 250 mg bid, 8 patients were given 500 mg were noted in the results of endocrine, hepatic, or hematological bid (one reduced to 375 mg dose because of nausea). investigations. In particular, The levels of total T1 and T4 and of free T1 and T4 did not alter. Otherwise probucol was well tolerated. "The probucol regimen was simple, convenient, well tolerated and did not interfere with growth and development in children or adolescents." 67 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment West, 1980259 2 children refused further No intention-to-treat analysis. medication after the first dose. 1 boy died of tercurrent infection 10 months after starting cholestryamine. Only 55% remained on treatment after 6 years; 48% after 8 years. Of the 25 children under age 10 who were started on cholestryamin, 14 or 67% remained on treatment at the end of the study period as compared with 1 of the 10 (13%) of the children over age 10 (p<0.05). Other drug therapies Baker, 198254 Otherwise probucol was well tolerated. "The probucol regimen was simple, convenient, well tolerated and did not interfere with growth and development in children or adolescents." 68 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Becker, 1992264 Long-term treatment of severe Open-label clinical study To determine the effects of sitosterol and bezafibrate 30 months familial hypercholesterolemia in in sequence for 3-month periods, and a combination exposure to drug: children: effect of sitosterol and Germany of both drugs at half-dose for 24 months, on serum 3 month diet bezafibrate lipid levels. Laboratory safety parameters, 3 month sitosterol ultrasonography, and physical examination assessed 3 month bezafibrate side effects. 24 months sitosterol plus bezafibrate Brun, 1980265 Effects of dextrothyroxine on the Descriptive To evaluate the effects of dextrothyroxine (D-T4) on 12 weeks pituitary-thyroid axis in Canada the pituitary-thyroid axis, the study measured the hypercholesterolemic children and secretion of TSH in response to TRH in six euthyroid goitrous adults children with FH. Since the effects of thyroid hormones appear to be mediated by specific nuclear receptors, the binding affinity of D-T4 was also studied. 69 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Becker, 1992264 1) family history of hypercholesterolemia and premature coronary heart 7 children with HeFH Mean age 8.42 disease in at least one first-degree family member 42.9% female 2) age 7 or older 3) total serum cholesterol >300 mg/dL and LDL >250 on at least 2 occasions with normal triglyceride levels after at least 3 months of strict dietary intervention 4) unsuccessful intervention with bile acid-binding resins 5) none of the children had phytosterolemia. Brun, 1980265 FH criteria: one of the parents had tendinous xanthomas with increased levels 6 Aged 7-12 of cholesterol in the LDL fraction. The children also had increased LDL. They 33% female were euthyroid and, due to their young age, had none of the clinical manifestations of FH except for hypercholesterolemia. 70 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Becker, 1992264 After 3-month diet that continued throughout the study, subjects After sitosterol: there was a slight but significant decrease in received sitosterol pastils (3x2 g/d) for 3 months, followed by hemoglobin (-5%) and alkaline phosphatase activity (-19%). bezafibrate (2x200 mg/d) for 3 months, followed by sitosterol After bezafibrate: alkaline phosphatase remained lower and iron (3x1 g/d) combined with bezafibrate (200 mg/d) for 24 months. increased by 26%. After 24 months of combined drugs: transferrin increased 20%, Safety was assessed by routine lab tests; abdominal sonography reaching abnormal levels in 2 patients. All other lab values remained performed every 6 months. within normal range. Decrease in appetite for 2 weeks was reported by 2 patients at the beginning of sitosterol therapy. Physical exams and ultrasounds of gallbladder found no abnormalities during follow-up. Brun, 1980265 All children were on a hypocholesterolemic diet at the time of the In the euthyroid hypercholesterolemic children before D-T4 treatment, study. At the first visit, TRH stimulation (400 ug, iv) test was the peak serum TSH response to TRH administration was normal. performed before and after 6 and 12 weeks of treatment with 6 With D-T4 treatment, the secretion of both TSH and T3 in response to mg D-T4 (choloxin). After 6 and weeks of therapy, the patient TRH was abolished. Also, an increase in the basal level of T3 was came back to the clinic for drug compliance assessment (via observed after treatment with D-T4. tablet counts), and measurements of serum T3 resin uptake and The high circulating levels of D-T4 and possibly of D-T3 after chronic total T4 levels. administration of D-T4 may be responsible for the saturation of pituitary nuclear T3 receptors, resulting in the suppression of the TRH-induced TSH response. 71 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Becker, 1992264 Children and parents reported a high degree of acceptance and compliance during all treatments. Brun, 1980265 NR 72 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Colletti, 1993266 Niacin treatment of Retrospective review of report of efficacy of niacin as single-drug treatment 1-19 months, hypercholesterolemia in children patients from 2 lipid clinics for hypercholesterolemia in children; average 8.1 +5.3 in Vermont (1986-1990) months and Boston (1980-1991) Malloy, 197873 Familial hypercholesterolemia in Single-blind random To determine the effect of p -aminosalicylic (PAS) on 1 year children: treatment with p- assignment pattern with serum cholesterol and triglycerides in 3 lipoprotein Aminosalicylic Acid cross-over at 6 months - classes in children and adolescents with severe drug familial hypercholesterolemia. USA in Children aged 5 to 21 years with severe familial hypercholesterolemia 73 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Colletti, 1993266 All children had been previously treated with diet and bile-acid sequestrants 21 children ages 4-14 Baseline values and had not reached NCEP goals; years (mean+s.d.): Niacin subsequently prescribed as sole treatment; this drug selected because TC=7.84+1.13 mmol/L of poor compliance (n=2) or anticipated poor compliance (n=11)with (303+44 mg/dL) sequenstrant therapy, low HDL (n=6) and parent preference (n=2). LDL=6.28+1.16 mmol/L (243+45 mg/dL) Malloy, 197873 1. Severe familial hypercholesterolemia 20 enrolled Median age: 12 years 2. Consistently elevated serum cholesterol while on > 4 month diet restricted in Female gender: 45% cholesterol and saturated fats 3. No disorders known to contribute to secondary hyperlipidemia 4. No other serious disorder 5. No medications other than PAS 6. One parent with proven and one other in kindred with hypercholesterolemia. 74 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Colletti, 1993266 Niacin as sole treatment for hypercholesterolemia; maximum one 7-year old girl developed a febrile influenza-like illness with daily dose was 7-98 mg/kg of body weight. elevated serum aminotransferase>400 IU/L two months after starting niacin treatment but before the first post-treatment lipid measurement. Pretreatment LFT's had been normal; Niacin at 1000mg/day sustained release was considered a possible cause of hepatitis and was discontinued. Dose reduced if nausea occurred. Flushing 71% itching 19% Abdominal pain 14% Nausea 14% headache 14% Constipation 5% Reversible serum aminotransferase elevations (dose related) in 6 patients; 4 with crystalline and 2 with sustained release form of niacin. 18 of 21 patients reported some adverse effect. Malloy, 197873 Dietary cholesterol and saturated fats restricted 4 months prior to No indicators of drug toxicity reported, other than mild gastric irritation and throughout the study. Subjects and their parents received that remitted with oral antacid treatment. No patient had occult blood monthly dietary counseling. Single-blind assignment to active in the stool at any time. No ophthamologic or hematologic drug (150 mg/kg/day, up to 8 gm/day) or placebo during first 6 abnormalities were observed. SGOT, SGPT, alkaline phosphatase, months. Dose adjusted monthly for change in body weight. After bilirubin, and glucose levels in fasting serum were consistently within 6 months, the alternate agent (placebo or active drug) was given normal limits in all subjects. There were no departures from expected for 6 months. Composite index of compliance for each subject growth curves during the administration of the active drug. Levels of based on interviews, tacit drug inventories, and tests for PAS TSH in serum were within normal range and showed no tendency to metabolites. Monthly clinic visits including clinical exam. rise during treatment with PAS-C. Thyroxine levels were normal in all samples. Triiodothyronine levels were normal in all but two in active drug and 1 on placebo. 75 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Colletti, 1993266 Drug discontinued in 1 pt for hepatitis; 1 for flushing; 1 for abdominal pain, 3 for vomiting; 2 for headache; 1 for elevation of serum aminotransferase, and 1 for poor compliance. Malloy, 197873 Compliance was >50% for 11 Single-blind randomization subjects, <50% for 9 subjects. No separate control group: patients served as own controls Only 11/20 subjects were > 50% compliant LDL change evaluated for 8 subjects only 76 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study McDuffie, Three-month tolerability of Orlistat Observational study, US To study the safety, tolerability, and potential efficacy 3 months 2002267 in adolescents with obesity-related of orlistat in adolescents with obesity and its comorbid conditions comorbid conditions. Stein, 1989268 Treatment of familial Cross-sectional To assess plasma lipid levels in children and N/A (cross-sectional hypercholesterolemia with drugs study/chart review adolescents who were on various drug regimens and study/chart review) in children Lipid clinic, US who were attending a specialized lipid treatment center. 77 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics McDuffie, Obese white and African-American adolescents (aged 12-17) were recruited 20 adolescents Mean age 14.6 2002267 through newspaper ads and letters to local physicians for participation in a 50% female weight-loss study. Inclusion criteria: BMI > NHANES I, 95th percentile for age, BMI 44.1 kg/m2 sex, and race; and the presence of one obesity-related comorbidity: hypertension, type 2 diabetes or glucose intolerance, hyperinsulinemia, hyperlipidemia, hepatic steatosis, or sleep apnea. Exclusions: a major pulmonary, hepatic, cardiac, or musculoskeletal disorder, history of substance abuse or other psychiatric disorder, use of a anorexiant in the past 6 months, or weight loss in the past 2 months. Stein, 1989268 Clinic charts from 30 HeFH children and adolescents who had been treated 30 HeFH children Mean age 5.5, range 1-20 with medication at the Cholesterol Treatment Center for 1 to 9 years were reviewed. 78 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects McDuffie, A registered dietitian instructed subjects and at least one parent All but one subject reported two adverse effects on at least one 2002267 or guardian how to follow a 500-kcal-deficit diet containing no occasion. Five of the 12 AEs queried: increased defecation, soft more than 30% of calories from fat. Subjects were given orlistat stools, fatty or oily stools, oily spotting on clothes, and increased flatus (120 mg, 3 times a day with meals) as well as a multivitamin were reported by >50% of subjects. AEs were generally mild, limited supplement. Subjects returned medication at 4-week intervals to to gastrointestinal effects related to increased fat excretion , and assess adherence. Subjects participated in a 12-week resolved within the first 6 weeks of treatment. comprehensive behavioral program that reinforced dietary A small but significant drop in 25-hydroxy vitamin D levels was seen at principles, encouraged physical activity, and provided 1 month. 3 subjects required additional vitamin D supplementation psychosocial support. despite the prescription of a daily multivitamin containing vitamin D. Subjects were queried weekly about AEs. At 3-month intervals, TSH, free thyroxine, glycosylated hemoglobin, calcium, phosphorous, a clinical pharmacist interviewed subjects with a comprehensive magnesium, zinc, measures of iron stores did not change significantly questionnaire to identify expected and unexpected AEs. during the study. Stein, 1989268 After stabilization and maximal lipid reduction by diet, drug The combination of resin and niacin produced elevated liver enzymes therapy consisted of a BABR, either colestipol or cholestyramine. (AST and ALT) and clinical symptoms of hepatotoxicity in one subject. Niacin treatment was added only when LDL lowering with resin Niacin in this subject was associated with a significant rise in liver alone was insufficient to decrease values below the 90th enzymes, suppression of albumin, and clinical symptoms of percentile. Three children with HoFH were also treated with a hepatotoxicity. No adverse clinical or biochemical side effects were variety of drug combinations, usually consisting of a BABR noted on the HMG CoA reductase inhibitor. (colestipol 15 to 30 g/day or cholestyramine 8 to 16 g/day), niacin (1 to 3 g/day), and in some cases, clofibrate. Six subjects with severe HeFH, where resin and niacin were inadequate, were entered into formal studies with an HMG CoA reductase inhibitor (lovastatin 80 mg/day or simvastatin 40 mg/day). 79 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment McDuffie, Subjects who completed 2002267 treatment (85%) reported taking 80% of prescribed medication. Only one subject (5%) cited intolerance of adverse effects as reason for withdrawal. Stein, 1989268 NR 80 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Steinmetz, Biological variations in Descriptive study To learn how effective fenofibrate is in treating 18 months 1981269 hyperlipidemic children and France hyperlipoproteinemia and whether it has secondary adolescents treated with effects on the biological parameters assessed in fenofibrate clinical laboratory tests. Wheeler, Double blind trial of bezafibrate in DB randomized cross-over To evaluate the use of benafibrate for lower 6 month crossover; 198582 familial hypercholesterolemia trial cholesterol in children. 3 months on bezafibrate Low-fat diet Cetta, 1994279 Growth patterns of hyperlipidemic Retrospective, descriptive To assess growth patterns of hyperlipidemic children 2 years of follow-up children enrolled in a preventive study enrolled in a preventive cardiovascular health clinic, cardiovascular health clinic Cardiovascular health by retrospective chart review. clinic, US 81 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Steinmetz, 17 patients were followed in the Diabetes Service. Most patients had 17 Aged 4-19 1981269 hyperlipoproteinemia Type Iia; four had mixed Type-Iib. Wheeler, Diagnosis of FH established by TC>6.7 mmol/l (269 mg/1000 ml), a type IIa 14 children; all had Aged 4-15 (mean 10.9) 198582 pattern on lipoprotein electrophorsesis and normal fasting TG (<1.5mmol/l) in previously been treated the pt; with either similar lipoprotein abnormalities in one of the parents or with dietary measures premature CHD death in a parent and a similar lipid abnormality in another (inadequate to control close relative. cholesterol), had been recommended cholestyramine but refused to take it. Low-fat diet Cetta, 1994279 Entrance criteria: 1) hyperlipidemia (LDL >110 mg/dL or TG >95th percentile 63 Mean age 7.8 (range 2 to with a normal LDL), 2) lack of any chronic disease, 3) a minimum of 2 years of 16) clinic participation. 47.6% female 82 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Steinmetz, All were put on a low-lipid diet (cholesterol <300 mg/day). After 3 The study found a significant and major decrease in total alkaline 1981269 6 months of diet, fenofibrate was given at 200 mg/day, phosphatase activities, uric acid, and bilirubin. 4 or 17 subjects had readjusting to 100 or 300 mg/day if needed. increases in ALT activity of at least 100%, and in AST activity of at Fasting blood samples were collected before treatment and after least 80%. These increases persisted as long as the treatment was 3 months, and after 6, 10, 12, 14, or 18 months in 10 patients. maintained. A decrease was observed only if the treatment was Reference values were obtained from a group of children of suspended. healthy families. Uric acid decreased significantly (mean change -20%). Bilirubin decreased significantly (mean change -19%). Inorganic phosphates decreased slightly but significantly. Albumin was unchanged. Alkaline phosphatase decreased by 15% on average. GGT activity decreased slightly (by 1.5 U/l). ALT increased by 9 U/l. AST increased by 11 U/l. Wheeler, 1. Children asked not to change their diets 1. One child had high alk phosphatase at the end of 3 months of 198582 2. 10-20 mg/kg/day given bid (in 100 mg tablets) given for 3 bezafibrate (had a slight intercurrent infection, and values returned to months normal) 2. Another child had slight transient rise in alanine transaminase during first 2 months on bezafibrate but values were normal by end of 3rd month 3. growth was satisfactory throughout trial 4. no other reports of adverse effects Low-fat diet Cetta, 1994279 All participants were counseled to eat an AHA Step-One Diet and All participants experienced normal height growth. Throughout the exercise regularly. Weight and height were measured 12, 30, follow-up period, no deviations from attained height growth curves and 48 weeks after the initial visit, and every 3 months thereafter. were observed. Two female patients experienced weight loss (at ages Dietary counseling, including review of 3 to 5-day diaries, was 10 and 17, respectively) that crossed below 2 major attained-weight performed every 6 weeks up to 1 year after the initial visit, and percentiles during clinic participation. One child had anorexia nervosa. subsequently every 3 to 6 months. The other child entered the clinic eating a parent-instituted "vegetarian diet" that was more restrictive than the prescribed AHA-Step-One Diet. 83 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Steinmetz, NR 1981269 Wheeler, In all but one case bezafibrate Rated poor quality 198582 was detected in the urine on all No results prior to cross-over. measurements during the active No wash out period between cross over drug phase, and in none of the periods. measurements while on placebo. All children declared a strong preference for this drug as compared with cholestyramine used previously. Low-fat diet Cetta, 1994279 Compliance with recommended exercise activities was not evaluated. Estimates of dietary compliance were performed based on diaries collected periodically throughout the study. 84 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Copperman, Nutrient quality of fat- and Cross-sectional case To assess the nutritional adequacy of low-fat, low- At least 1 year 1995270 cholesterol-modified diets of comparison tertiary care saturated fat, low-cholesterol-modified diets of children with hyperlipidemia ambulatory pediatric children with hyperlipidemia. atherosclerosis prevention center DISC Efficacy and safety of lowering RCT To assess the efficacy and safety of lowering dietary 3 years Collaborative dietary intake of fat and 6 centers intake of total fat, saturated fat, and cholesterol to Research cholesterol in children with Prepubertal boys and girls decrease LDL-C levels in children Group, 1995219 elevated low-density lipoprotein recruited from public and cholesterol: The dietary private elementary intervention study in children schools, by mass mailings (DISC) to members of an HMO, and from pediatric practices Feoli-Fonseca, Familial lipoprotein lipase Retrospective, descriptive To describe the clinical and biochemical features of 23 years reviewed; 1998282 deficiency in infancy: clinical, study, Quebec, Canada familial LPL deficiency presenting in infancy, to mean duration of biochemical, and molecular study evaluate the safety and efficacy of severe dietary fat follow-up per patient restriction, and to define the LPL gene defects was 6.76 years responsible for the disease. 85 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Copperman, All children in the study were from white middle-class families from suburban 54 Mean age 10.8 1995270 New York City area. Hyperlipidemic children were referred to the Schneider 54% female Children's Hospital Center for Atherosclerosis Prevention for evaluation and treatment. Controls consisted of 4th and 5th-grade children from a local elementary school recruited from 2 classes. These children were free of chronic illness and were consuming unrestricted diets. DISC 1. Girls aged 7 years, 10 months - 10 years 1 month; boys aged 8 years, 7 663 (362 boys, 301 girls) Mean age: boys 9.7 Collaborative months - 10 years, 10 months girls 9.0 Research 2. Average of 2 screening LDL-C values > 80th and < 90th percentile for age Group, 1995219 and sex. 3. No medication or medical condition that could effect growth or blood cholesterol 4. No behavior problems in child or family likely to reduce adherence 5. Prepubertal 6. No plans to move within the 3 study years Feoli-Fonseca, Charts were reviewed for all patients with chylomicronemia who attended the 16 All patients were diagnosed 1998282 lipid clinic of Hospital Sainte-Justine between 1972 and 1995. Inclusion at age <1 year. criteria: presentation before 1 year of age, French-Canadian ancestry, and 62.5% female initial plasma TG > 10 mmol/L. 18 were selected; 16 were found to have LPL enzyme deficiency. To assess safety, the authors compiled available weight, height, and sex maturation stage data during the follow-up period, and hemoglobin levels and blood chemistry values of iron, alkaline phosphatase, total calcium, red blood cell folate, and albumin. 86 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Copperman, Subjects received individual nutrition counseling on a NCEP-Step There was no significant difference in consumption of energy, 1995270 1 diet from a registered dietitian. Subjects and controls attended minerals, or vitamins D and E between the groups. The children with separate training sessions in which a registered dietitian hyperlipidemia consumed significantly more vitamin A (p<0.005). instructed the children and/or families on the accurate completion of the food record and a food frequency questionnaire. DISC Randomized to diet (334) or usual care (329). No differences between groups in adjusted mean height, serum ferritin Collaborative Diet: Behavioral intervention to promote adherence to a diet levels, or other safety measures. Research providing 28% of energy from total fat, less than 8% from Serum ferritin decreased in both groups and the intervention group had Group, 1995219 saturated fat, up to 9% from polyunsaturated fat, and less than slightly lower mean ferritin concentrations than did usual care at year 3 75/mg (1000 kcal) per day of cholesterol (<150 mg/day). Family (p=0.08). In both groups, mean ferritin concentrations remained above oriented, based on social learning and social action theory. 1st the 75th percentile for age and sex. visit: eating pattern assessed and personalized program Risk of consuming less than 2/3 of the RDA was significant for vitamin developed. In the 1st 6 months: 6 weekly and then 5 biweekly E at all visits (baseline OR: 1.009; year 1 OR: 1.007; year 3 OR: 1.007; group sessions augmented by 2 individual visits of children with p<0.0001 for all visits); for zinc at all visits for boys (baseline OR: their family members. In the 2nd 6 months, 4 group and 2 1.004, p<0.05; year 1OR: 1.003, p<0.02, and year 3 OR: 1.004, individual sessions. Years 2 and 3: group individual and p<0.003) and girls (baseline OR:1.007, p<0.001: year 1 OR: 1.008, p< maintenance sessions held 4-6 x per year with monthly phone 0.0006 and year 3 OR: 1.005, p < 0.003). contacts between sessions. Usual care: Public educational publications on heart-healthy eating provided. Parents informed if child blood cholesterol high - no specific recommendations to see physician given. 3-year lipid results provided, with referral as clinically warranted. Feoli-Fonseca, Dietary intervention was started in infancy at the hospital. 2 children had abdominal pain requiring hospitalization, and another 1998282 Patients were first given an electrolyte solution intravenously or had a documented episode of acute pancreatitis 2 months after the orally. Fats were progressively reintroduced to reach a level of introduction of an oral contraceptive agent. 5% to 7% of energy as long-chain TG. Solid foods were added No persistent adverse effects on growth were seen. Abnormal values after 4 months. At 1 year of age the special formula was were observed for serum iron, alkaline phosphatase, and total calcium. replaced by skim milk, and lean protein foods progressively All erythrocyte folate (15 measurements, 8 patients) and albumin (41 added. Diet provided at most 10% of energy as long-chain TG. measurements, 11 patients) levels were within the normal range. 87 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Copperman, NR 1995270 DISC Attendance at intervention Rated good quality Collaborative sessions averaged 96% during Research the first 6 months, 91% during Group, 1995219 the second 6 months, 91% during the second year, and 89% during the third year. Feoli-Fonseca, NR 1998282 88 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Jacobson, Normal growth in high-risk 3-year longitudinal To assess the safety and efficacy of lowering dietary 3 years 1998280 hyperlipidemic children and descriptive study intake of total fat, saturated fat, and cholesterol in adolescents with dietary growing children and adolescents with severe intervention hyperlipidemia. Kaistha, Overrestriction of dietary fat intake Cross-sectional study To assess the nutritional adequacy of the diets of N/A: cross-section 2001285 before formal nutritional children with hyperlipidemia following medically counseling in children with unsupervised low-fat diets compared with children hyperlipidemia receiving unrestricted diets. Kuehl, 1993227 Effective control of RCT of parental and family To identify and characterize a family-based group 33 weeks hypercholesterolemia in children education intervention to lower plasma cholesterol in children. with dietary interventions based in USA in children with TC pediatric practice. >185 ages 2-15 (mean 7) 89 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Jacobson, All children were diagnosed with hyperlipidemia and had been referred by their 138 Age range 2-15 1998280 pediatricians. Patients who were older than age 15 at the first visit were 46% female excluded in order to minimize the effects of cessation of growth follow gin 84.1% white, 5.9% African- puberty. Subjects with secondary hyperlipidemia caused by hepatic, renal, or American, 8.7% Hispanic, thyroid disease were also excluded. TC > 95th percentile for age, and had at 1.5% other race least 3 visits over 3 years. Follow-up visits were at 3- to 6- month intervals. Kaistha, Children with hyperlipidemia were selected from patients referred Apr 1, 1996 46 c/ hyperlipidemia Mean age 9.7 2001285 to Dec. 31, 1997 to the Children's Cardiovascular Health Center (CCHC) at 34 healthy controls 47% female Columbia-Presbyterian Medical Center. 46 were found to be meeting current 57% Latino NCEP Step I diet recommendations for total fat (<=30% of calories). 43% white Exclusions: TG ?400 mg/dL or homozygous LDL-C receptor deficiency. Controls were recruited from local pediatric practices in the vicinity of the Medical Center. Controls were invited if their parents stated they were healthy, were free of chronic illness, and were not following a modified or restricted diet. Kuehl, 1993227 Children referred by physicians who participated in a seminar on the risk 295 children randomized SSI group: factors for atherosclerosis in children, identification and management of high to either a single session mean age 6.6 cholesterol in children. 14 pediatric practices referred 295 children. intervention (SSI) or a gender: NR multi-session intervention race: NR (MSI) MSI group: mean age 7.6 gender NR race NR 90 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Jacobson, Diet restricting total fat content to 30% of total calories, and The participants grew as expected during the follow-up period. There 1998280 saturated fat to 10% of total calories. Anthropometric measures, was no significant change in height or weight percentile, expressed as lipid profiles, and dietary assessment were obtained at each visit. Z score, from baseline to 3-year follow-up. Kaistha, At baseline, children and parents were given a 3-day food record A significantly (p<0.05) greater proportion of children with 2001285 with detailed instructions and training by a registered dietitian. hyperlipidemia consumed below 75% of the RDA/DRI for Vitamin E Children were asked to complete 3 days of food records, and calories, compared with controls. 90% of the decrease in caloric including 2 weekdays and 1 weekend day. No dietary counseling intake could be accounted for by a decrease in total fat intake. or guidance was given to children or parents at the time of diet- Many children in both groups also did not meet the RDA for folate and monitoring instruction. Anthropometric data were obtained at calcium, but between-group differences were not significant. baseline, and nutrient intakes were analyzed. Kuehl, 1993227 1. SSI=one 90 minute nutrition education session for patient, Both groups maintained iron intake at over 87% and calcium intake at siblings and parents (slides presentation, low fat food prep and over 81% of the RDA throughout the study. % RDA of iron reported tasting, distribution of fruit and cereal). "Returned for baseline to visit 2: MSI 103+/-5 to 98+/-5 (pNS); SSI 87+/-4 to 90+/-4 subsequent food sampling with nutritionist present" but no further (p=NS). formal education. 2. MSI= four 90 minute sessions with focus on food preparation All patients "grew linearly with a mean height increase of 2.0+/-0.3 for breakfast, snack, lunch and dinner. Received notebooks with inches in the MSI group and 2.0+/-0.1 in the SSI group. Both groups nutrition information and recipes, incentives for attendance and had a mean weight increase of 4.1 lbs. Growth parameters not completion of behavioral contracts (eating low fat meal) significantly different between the two groups. SS outcomes were measured at baseline, 8.5 weeks and 21 weeks MSI outcomes were measured at baseline, 9 weeks and 33 weeks 91 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Jacobson, NR 1998280 Kaistha, N/A: cross-section Authors comment that the findings 2001285 suggest that without formal nutritional counseling, parents of children with hypercholesterolemia may inadvertently overrestrict calories in their children's diet by attempting to eliminate obvious sources of dietary fat. Kuehl, 1993227 "Compliance with providing food Poor quality: records was excellent during the Large and unequal drop out rates (35% intevervention." for SSI and 16% for MSI) and analyses only use those who completed the second blood draw (215 of original 295); ie not intention to treat. No assessment of intervention fidelity. 18% of MSI and 20% of SSI had plasma TC values <170 and were excluded from the analyses because they didn't meet the initial referral criteria of TC >185. 92 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Lavigne, A cholesterol-lowering diet does RCT To describe the 3-year psychosocial safety results of Mean length of 1999281 not produce adverse 6 centers the DISC study, in which a wide range of possible follow-up 36.2 psychological effects in children: Prepubertal boys and girls behavior and cognitive problems were assessed. months three-year results from the dietary recruited from public and intervention study in children private elementary schools, by mass mailings to members of an HMO, and from pediatric practices Lifshitz, 1989278 Growth failure - a complication of Observational study, US To describe 8 patients with growth failure among a Mean time from dx dietary treatment of group of 40 children who were advised by their to nutritional hypercholesterolemia pediatricians to eat a low-fat, low-cholesterol diet for consultation: 20.1 the treatment of hypercholesterolemia. They mos. for patients consumed inappropriate diets with insufficient energy with growth failure; and micronutrients to sustain normal growth and 3.9 mos. for weight gain. patients without growth failure (p<0.001) 93 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Lavigne, Children were eligible if the average of 2 LDL screening values was between 663 Aged 8-10 1999281 80-98th percentile, aged 8-10, and prepubescent (Tanner Stage 1). 47% female Exclusions were taking medication that affected growth or cholesterol; onset of puberty, non-English speaking; in a remedial special education class; high parental alcohol consumption; or total behavior problem score on CBCL >98th percentile. Lifshitz, 1989278 40 children were referred in 1986-1987 to the Nutrition Center, Division of 40 87.5% Tanner Stage I pediatric endocrinology, North Shore University Hospital in NY because of 10% Stage II hypercholesterolemia. The children had no other medical abnormalities. 2.5% Stage III Mean age at diagnosis 7.7 94 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Lavigne, A total of 663 8- to 10-year-old children with elevated LDL were 3-year results showed no adverse effects of treatment in academic 1999281 randomly assigned to either an intervention or a usual-care function, psychological symptoms, or family function. When group. Intervention included group and individual counseling statistically significant effects were observed, they generally indicated a sessions to assist participants in adopting a dietary pattern beneficial effect from the dietary intervention. containing 28% or less of calories from total fat (<8% saturated At 3 years, there were no significant treatment group differences in fat, up to 9% polyunsaturated fat, and 11% as monounsaturated. CBCL total behavior problem, externalizing, or internalizing scores. fat) and dietary cholesterol intake of less than 75 mg/1000 kcal. However, the intervention group scored significantly lower on CBCL Psychological measures included academic achievement (W-J), Thought Problems and Social Competence subscales. Intervention- family cohesion or conflict (FES), total behavior problems group girls had lower CBCL Anxious-Depressed subscale scores than (CBCL), diagnosed eating disorders and suicide threats, trait usual-care-group girls (-0.029, p<0.05). There were no treatment anxiety (STAIC), or depression (CDI). group differences in CBCL Aggression or Conduct Problem subscale scores. Self-reports of anxiety showed no differences between groups, but self-reports of depression were significantly lower for the intervention group than for the usual-care group (-0.75, p<0.05) Lifshitz, 1989278 The referring physician gave general advice to reduce levels of 8 (20%) of 40 patients had growth failure. Three of the 8 had dietary fat and cholesterol at the time of diagnosis. The specific nutritional dwarfing according to the anthropometric indexes of the recommendations varied in each case, and implementation by Welcome Trust Classification, with no progression of puberty. The the families was unsupervised. The patients' dietary intake and other 5 had a drop in body weight without linear growth alterations. food patterns were evaluated the Nutrition Center 2 weeks to 4.5 Total dietary energy and zinc content of patients with growth failure years after diagnosis. Weight and height records throughout life were significantly less than those with normal growth. The diets of were plotted on standard growth charts. Body composition was patients with growth failure provided an average of 66% of energy assessed by arm measurements as described by Frisancho. requirements and 40% of daily zinc requirements for their ideal body weight, age, and sex. Patients with normal growth consumed 84% of the energy requirements, and 58% of daily zinc requirements (p<0.05). Patients in both groups had similar fat intake: 25.4% with growth failure, 29.6% with normal growth; cholesterol and protein intake were also similar and appropriate (14.6% to 15.9%). The 3 patients with nutritional dwarfing had inadequate intake of vitamins and minerals: <50% RDA of vitamins B6, D, folacin, zinc, iron, calcium, and magnesium, and <66% of RDA for vitamins A and B12, niacin, and phosphorus. Only vitamins E and C were adequately consu 95 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Lavigne, Compliance not reported. A 1999281 total of 96% of the intervention and 92% of the usual-care participants returned for the 3- year follow-up visit. Lifshitz, 1989278 NA 96 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study McKenzie, Change in nutrient intakes, Prospective cohort study, To determine change in nutrient intakes, number of 3 months 1996274 number of servings, and Pennsylvania, US servings, and contributions of total fat from food contributions of total fat from food groups in children who lowered their dietary fat groups in 4- to 10-year-old intake. children enrolled in a nutrition education study Moreno, Lymphocyte T Subset Counts in Descriptive study To investigate the immunological effects of a step 1 6 months 1998275 Children with diet from the NCEP in children and adolescents with hypercholesterolemia receiving hypercholesterolemia. dietary therapy 97 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics McKenzie, The CHP was a prospective cohort study (1991-1994) with three groups of 300 children from aged 4-10 1996274 children aged 4-10 with hypercholesterolemia (LDL between 80th-98th suburbs north of percentiles) and one age- and gender-matched group of children with normal Philadelphia, PA cholesterol. All children were recruited via their pediatricians' offices in suburbs north of Philadelphia. Moreno, Hypercholesterolemic children were recruited from 3 groups: children whose 42 Aged 7-15.9 years 1998275 parents presented dyslipoproteinemia; children whose parents suffered from early ischemic heart disease, and children with hypercholesterolemia detected by routine analyses. Excluded immunological diseases, use of corticosteroids or immunosuppressive drugs in previous month, hepatobiliary diseases and/or secondary dyslipoproteinemias. 98 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects McKenzie, Participants were assigned randomly to one of four study groups. Regardless of study group, 74% of all the children failed to consume 1996274 2 groups of hypercholesterolemic children received different two thirds of the RDA for vitamin D. Children in this study had on forms of nutrition education: One group of children and their average adequate intakes of all other nutrients, when they reduced caregivers received face-to-face counseling with a registered their fat consumption. dietitian and had free telephone access to the dietitian during the study. The other group was provided with an at-home parenting autotutorial program (PCAT). Both programs promoted NCEP recommendations: total fat <30% of energy, <10% saturated fat; cholesterol <100 mg/1000 kcalories. The third group of hypercholesterolemic children and the control group received no formal nutrition education. Three 24-hour dietary recalls were collected by telephone interview at baseline and 3 months later, and measured for nutrient and food group intakes. Moreno, Patients and at least one parent met with a pediatric nutritionist Lymphocyte T subset counts (CD3, CD4, and CD8) showed significant 1998275 who established recommendations based on NCEP Step 1 diet decreases after 6 months of dietary therapy. In all cases, lymphocyte (total fat 30%, saturated fat 10% of total energy). Detailed 24-hr T subset counts remained within normal ranges. Changes in CD3 and dietary recalls were obtained at baseline, and 6 months after CD8 counts were significantly correlated with changes in triglyceride beginning therapy. Physical activity habits were not modified serum levels (p<0.05). The other immune indexes (immunoglobulins during the study. At baseline and 6 months, fasting blood was G, A, M, and complements C3, C4, and Factor B) did not change analyzed for lipoprotein profile, immunoproteins, and lymphocyte significantly. T subsets. 99 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment McKenzie, NR 1996274 Moreno, NR 1998275 100 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Obarzanek, Long-term safety and efficacy of a RCT To evaluate the long-term safety and efficacy of a 4 years 2001218 cholesterol-lowering diet in Six clinical centers in the cholesterol-lowering diet in children with elevated children with elevated low-density USA LDL lipoprotein cholesterol: seven-year results of the dietary intervention study in children (DISC) Rose, 1976277 Primary hyperlipoproteinemia in Descriptive study To identify primary and familial hypercholesterolemia 16 patients followed childhood and adolescence: Toronto, Canada in children and adolescents; perform dietary for 1 year; identification and treatment of treatment and continued assessment of effects and 9 patients followed persons at risk for premature side effects of therapy. > 1 year atherosclerosis 101 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Obarzanek, 1. Average of 2 LDL-C values in 80th - 98th 663 randomized, Intervention group 2001218 sex-specific percentiles for 8 - 10 year old children in a fasting state 334 to intervention Mean age: 9.5 (.74) 2. No medical conditions or medications that could effect growth or serum 329 to control Female gender: 46% cholesterol, behavioral problems or onset of pubertal maturation. Race: 86.5% of entire study sample white Control group Mean age: 9.5 (.70) Female gender: 44% Race: 86.5% of entire study sample white Rose, 1976277 The populations screened were 936 patients at pediatricians' offices, 1232 4013 were screened; 16 Age range: 1 month to 20 inpatients without conditions likely to cause secondary hyperlipidemia, and cases of primary years 1845 patients with congenital heart disease attending the outpatient clinic or hyperbetalipoproteinemia being investigated in the cardiac catherization lab. Betalipoprotein cholesterol were identified was quantified in serum from subjects in whom the initial diagnosis was primary hypercholesterolemia and from children with borderline abnormal cholesterol values on repeat examination. Secondary causes of hypercholesterolemia such as diabetes, renal or liver disease, hypothyroidism and dysglobulinemia were excluded. Parents and siblings of affected individuals were screened to assess whether the primary lipid abnormality was familial. 102 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Obarzanek, Randomized to dietary intervention or control group. No differences at any data collection point in height or serum ferritin or 2001218 any differences in an adverse direction in red blood cell folate, serum Intervention: Dietary behavioral intervention that promoted retinol and zinc, sexual maturation or body mass index. adherence to a diet with 28% of energy from total fat,< 8% from saturated fat, up to 9% from polyunsaturated fat, and <75 mg/1000 kcal cholesterol per day. Control: Usual care - parents informed child's blood cholesterol high and were given educational materials on heart-healthy eating as available to the public. Annual examinations for study- wide measurements. Intervention: 1st six months: 6 weekly and then 5 biweekly group sessions led by nutritionists and behaviorists, and 2 individual sessions with nutritionist. 2nd six months: 4 group and 2 individual sessions. 2nd and 3rd years: group and individual maintenance sessions held 4 - 6 x/year, with monthly phone contacts between group sessions. 4th year: individualized approach with motivational interviewing and stage of change. 2 group events and 2 individual visits annually, with individual phone contacts "as appropriate". Rose, 1976277 A diet low in cholesterol and high in polyunsaturated fats was Growth and development of these patients has progressed according instituted. to growth percentiles and no side effects of therapy have been detected. 103 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Obarzanek, Intervention attendance 2001218 diminished from an average of 96% in the first 6 months to 89% during year 3, and 72% during year 5. After year 5, in the last 3 years of intervention, the percentage of intervention children having 2 or more visits per year decreased from 55% to 42% to 37%, and on average 1.8, 1.4, and 1.3 intervention contacts per participant per year were achieved. Rose, 1976277 NR 104 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Sanchez-Bayle, Diet therapy for Observational To evaluate the effectiveness and compliance of 6-24 months (29 for 1994276 hypercholesterolemia in children Spain dietary restriction in a group of children and 2 years; 160 for 1 and adolescents adolescents with hypercholesterolemia, and monitor year; 262 for 6 growth velocity. months). Sanchez-Bayle, Influence of dietary intervention Prospective descriptive To determine whether a moderately reduced fat diet Mean 7.42 years 2003283 on growth in children with study, Spain affects longitudinal growth in children with hypercholesterolemia hypercholesterolemia Segall, 1970272 Effects of short-term high- Observational To described the effects of high carbohydrate diet on 10 days carbohydrate feeding on serum London triglycerides in children with FH. triglyceride of children with familial hypercholesterolemia 105 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Sanchez-Bayle, 507 children referred to Nino Jesus Children Hospital after school-based 451 children and 211 males (46.8%), 240 1994276 screening showed a total cholesterol value of 200 mg/dL or higher. More adolescents 2-18 years females (53.2%) complete lipid study was performed after fasting, and those with cholesterol persisting above 200 mg/dL and LDL above 130 mg/dL were included. Of these 471 meeting criteria, 18 did not show up for 1st visit, 38 were unavailable for f/u. Sanchez-Bayle, Children who had a TC > 200 mg/dL and LDL > 135 were included. Children 144 children in Madrid, Age 2-13 y, mean age 5.53 2003283 with chronic disease were excluded. Spain at beginning of treatment 52% female Segall, 1970272 Unclear 5 children with HeFH (6 age range 6-15, 4 female total but results presented and 1 male here without his data). 106 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Sanchez-Bayle, Dietary treatment was according to the AHA Step-One Diet Growth velocity was found to be normal in all subjects. 1994276 proceeding to Step-Two Diet when TC persisted above 240 mg/dL after 5 months. Dietary counseling was given by one pediatrician to both parents and the child. Detailed written recommendations were given to each family. All subjects were advised as to the necessity of aerobic exercise and dangers of tobacco Sanchez-Bayle, AHA Step-1 diet was implemented, proceeding to Step 2 if TC > Change in Z-score for height was not affected by moderate dietary fat 2003283 240 persisted after 5 months, as well as OMS 1985 and 10th restriction (0.58 baseline to 0.45 endpoint). Subjects had a significant RDA recommendations for caloric intake by age. Dietary increase in weight (p<0.001), from Z xcore of 0.33 at the beginning of counseling was given to both parents and children by the the study, to 0.58 at endpoint. The authors conclude that growth is not pediatrician. Compliance was assessed at 1 month, 6 months, influenced by moderate fat restriction. and each year using a questionnaire. Patient weight and height were measured each year and compared with growth standards for Spanish children. Segall, 1970272 10 day high carbohydrate diet: 75% of total calories from 1 child lost 1.4 kg. carbohydrate (of which 75% sucrose); 10% from fat and 15% from protein. Diet was given as normal foods and was isocaloric with home diet prior to intervention. Only 3 of the 5 children had blood levels measured after 10 days. 107 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Sanchez-Bayle, Follow-up was 100% at 6 1994276 months; 42% at 1 year and 6% at 2 years. Authors report 69% compliance with diet at 1 year, but data presented for 189 (42%) participants. Sanchez-Bayle, Assessed but not reported 2003283 Segall, 1970272 NR 108 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Tershakovec, Growth of hypercholesterolemic RCT of dietary intervention To evaluate the growth of hypercholesterolemic 12 months 1998284 children completing physician- USA children completing a physician-initiated home-based initiated low-fat dietary nutrition education program or standard nutrition intervention counseling to lower dietary fat intake. (PCAT) Tonstad, Psychosocial function during Cross-sectional study To determine whether children treated for FH have a N/A (cross- 1996271 treatment for familial Norway greater psychosocial burden than their peers. sectional); children hypercholesterolemia treated for 18 mo- =9 yrs prior to this study 109 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Tershakovec, 1. Mean fasting plasma LDL-C 107-164 mg/dl for boys and 112-164 mg/dl for 261 children with Children aged 3.9 - 9.9 years 1998284 girls hypercholesterolemia recruited from 9 suburban 2. Free of secondary causes of dyslipidemia recruited from 9 suburban pediatric practices 3. At least 85% but not greater than 130% of ideal body weight pediatric practices, PCAT group randomized to home- Mean age: 6.3 + 0.2 based education (PCAT) Female gender: 51% standard nutrition Counseling group counseling Mean age: 6.2 + 0.2 or at-risk control group. Female gender: 50 Not-at-risk control group At-risk control group randomly selected from Mean age: 6.4 + 0.2 children whose TC was Female gender: 48% not elevated Not-at-risk group Mean age: 6.4 + 0.2 Female gender: 51% Race of all participants stated as 84 - 99% white. Tonstad, FH diagnosis was made if the child had a TC >259 mg/dL, and if one or both 185 children from 156 44.9% female 1996271 parents and other relatives had TC >300 mg/dL or tendon xanthomas or if the families Aged 7-16 LDL receptor mutation was identified (this was the case in 77%). 5 were excluded for familial combined hyperlipidemia, 3 for HoFH, 1 for a disabling chronic disease, and 1 for mental retardation. 110 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Tershakovec, PCAT and counseling groups complied with recommendations of None noted. Outcomes included 1998284 the National Cholesterol Education Program Expert Panel on 1. Height Blood Cholesterol Levels in Children and Adolescents. 2. Weight PCAT: 10 talking books lessons and follow-up pencil/paper 3. Skinfold measures activities for children with manual for parents based on social 4. Dietary intake cognitive theory. A story and activity completed each week for 10 weeks. Percent calories from fat was associated with weight z-score (p<0.01), Counseling: Child and at least 1 parent had 45-60 minute weight for height median (p<0.005), and sum of skinfolds (p<0.05). session with a pediatric registered dietitian. Caloric intake was positively associated with weight (p<0.005) and Controls: Not provided educational information or materials height (p<0.005) z-scores and weight for height median (p<0.01). After controlling for age, no significant time-related among-group differences Measures evaluated at baseline, 3, 6 and 12 months were observed for the height or weight z-scores, weight-for-height median, caloric intake, or fat intake. Differences among groups in sum of skinfolds approached significance (p=0.06). Authors note results analysis shows these children could consume a relatively low-fat diet (down to 24.3% calories as fat in the lowest fat intake quintile group) and still maintain growth over the subsequent year. Tonstad, Mean years of treatment: 4 (range 1.5-9). Recommended diet Overall behavioral and emotional scores of children with FH were 1996271 was <30% total dietary energy from fat, <10% of total energy similar to scores of a population-based sample. from saturated fat, and <100 mg cholesterol/1000 kcal. About Mean CBCL scores and TRF scores for FH-children were similar to the 2/3 of children had been treated with BABRs at some point. reference group. YSR total and externalizing scores were lower in girls Dietary compliance during the past year was assessed by food with FH than in the population. Total and content area CAS scores frequency questionnaire. Psychosocial assessments included were similar for children with FH and the epidemiologic cohort, except the CBCL, the Teacher's Report Form, the Child Assessment for the scores for family, mood, and expression of anger, which were Schedule, and the Children's Global Assessment Score. For a lower in the FH group. reference group, 2600 children aged 4-16 were sampled from According to the CAS item, none of the children with FH had weight the Norway Central Population Register and mailed the CBCL loss or gain indicating the presence of an eating disorder. and TRF. 111 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Tershakovec, NR Participants were from predominantly 1998284 white and relatively affluent families; results may not generalize well. Diet evaluated by three 24-hour dietary recalls per assessment period Tonstad, Results of the FFQ indicated Assessments were made by parents and 1996271 good compliance with the diet, teachers. except for a slight increase in saturated fat intake among girls with FH. 112 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Witschi, Family cooperation and Observational, To study the effect of a family dietary program for the 4 week baseline 1978273 effectiveness in a cholesterol- uncontrolled study purpose of lowering serum cholesterol period followed by 3 lowering diet Boston, MA week diet change period Dietary supplements Amundsen, Plant sterol ester-enriched spread Randomized, double-blind To assess effect of SE-enriched spread on serum Study period: 25 200241 lowers plasma total and LDL crossover lipids, lipoproteins, carotenoids, fat-soluble vitamins, weeks with run-ins cholesterol in children with familial Subjects recruited from and physiologic variables in children with FH aged 7 - and washouts hypercholesterolemia patient register at Lipid 12 years. 2 - 8 week Clinic of National Hospital treatment periods. in Oslo, Norway 113 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Witschi, families recruited by notices in the academic community, telephone 91 adolescents 55% female, age NR 1978273 prescreening in the community and letters to volunteers for a prior study. Eligible if 1) an adult and at least one adolescent all living and eating most of their meals at home. Dietary supplements Amundsen, Not specifically noted. Implied: 38 Mean age: 10.5 + 1.7 200241 1. Parent with hypercholesterolemia 2 non-study-related drop- 50% female 2. Diagnosed with "definite" or "possible" heterozygous FH outs; 1 drop out because Cholesterol (mmol/L) 3. Healthy, with no clinical symptoms of hypercholesterolemia the amount of spread total: 7.01 + 1.26 required to be consumed LDL: 5.39 + 1.42 was too large. HDL: 1.37 + 0.32 Triacyglycerol (mmol/L) 0.56 + 0.22 114 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Witschi, A food record was kept by each participants for two weeks during No significant changes in weight. 1978273 4-week baseline period. Each family was assigned a nutritionist who maintained contact throughout the study. The diet was based on AHA guidelines for decreasing dietary cholesterol and saturated fat. Weight and blood samples were taken at beginning and end of baseline period, and days 10 and 21 of the 3-week diet-change period. Samples were assayed for cholesterol analysis. Dietary supplements Amundsen, Exposure: ALT increased in SE diet 16.8% (p = 0.04). No subjects had plasma 200241 In a double-blind crossover, with two 8-week interventions, 38 concentration of ALT outside the reference concentration at any time. children with FH consumed 18.2 + 1.5 g SE spread/day, Starting concentration of ALT was significantly lower in the SE diet corresponding to 1.60 + 0.13 g SE, or a control spread. Blood period compared to control. samples were analyzed at the start and end of each diet period. Serum levels of lycopene and B-carotene decreased with SE. After Subjects consumed a recommended American Heart adjustment for lipid changes, differences in B-carotene disappeared; Association Step I diet during both intervention periods. lycopene concentration was still 8.1% lower (p=0.015) in SE compared to control. Serum levels of retinol were higher in SE, and after lipid standardization: 15.6% (p<0.001) higher than control. After lipid standardization, ) a tocopherol was 7.1% higher in SE than in control (p=0.027). Authors note results may be influenced by different starting concentrations in SE than control. 115 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Witschi, Compliance was better among 1978273 females than males. More than 60% of the families attained an adherence rating of "excellent" or "good" during the diet test. Dietary supplements Amundsen, Children consumed 90.9% of Small N 200241 the control spread and 91.7% of Many children used supplements or the SE spread. 32% of subjects vitamins during the study. felt that the amount of spread These variables were not analyzed was too large, but 68% were separately. satisfied with the amounts. 46% FH was confirmed by mutation analysis of subjects could recognize a in only 25/38 subjects. difference between the 2 Some included children may not have spreads but could not identify had FH, as baseline total and LDL which was SE, of whom 56% cholesterol concentrations were indicated that the SE spread significantly higher in the confirmed FH tasted better than the control group. spread. 116 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Amundsen, Long-term compliance and Open-label trial, following To look at any sustained long-term effect of plasma 26 weeks 2004287 changes in plasma lipids, plant RCT above lipids and level of compliance following an sterols, and carotenoids in uncontrolled free intake of plant sterol ester-enriched children and parents with FH spread; to measure changes in serum consuming plant sterol ester- concentrations of non-cholesterol sterols, fat-soluble enriched spread vitamins and carotenoids in children and their parents during the extended open-label period. Becker, 1993 Treatment of severe familial Open-label clinical study To compare the ability of two plant sterols to reduce 10 months: 288 hypercholesterolemia in childhood serum lipid levels and to compare their mechanism 3 on sitosterol with sitosterol and sitostanol Germany of action in children with severe HeFH. 7 on sitostanol 117 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Amundsen, 41 healthy children were recruited from the patient register at The Lipid Clinic 37 mean 9.6 2004287 at the National Hospital in Oslo. All subjects were diagnosed with 'definite' or 'possible' heterozygous FH. The diagnosis was documented by the presence of an FH mutation in 25 of the children. In the other 16 subjects, the responsible mutation was still unidentified. All children were healthy and none had clinical symptoms of hypercholesterolemia or used serum cholesterol- lowering drugs. Becker, 1993 1) family history of hypercholesterolemia and premature coronary artery 9 children with HeFH Mean age 11.96 288 disease in at least one first-degree family member, 33.3% female 2) serum cholesterol concentration (TC and LDL) that was on two occasions 1.5 times >95th percentile for age and gender after at least 3 months of strict diet, 3) exclusion of phytosterolemia, a rare inborn error of hyperabsorption of plant sterols. 118 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Amundsen, 37 children were advised to eat plant sterol ester spread (20 g of lipid-adjusted lathosterol levels were stable; sitosterol levels were 77% 2004287 spread per day, corresponding to 1.76 g/day of plant sterol). higher compared to control period of RCT. Blood samples were taken at weeks 1 and 26. A total of 13 Lipid-adjusted retinol levels were 11% higher at the end of the open children took fish oil supplements (5 ml/day) and eight children label period as compared to control. No changes in lycopene or lutein. took multivitamin supplements during the study. lipid-adjusted serum a-carotene increased in the open label period following a decrease in the trial period. Becker, 1993 After a 3-month diet, subjects were given sitosterol pastils (3x2 After dietary intervention, alanine aminotransferase was decreased. 288 g/d) for 3 months, followed by a 7-month course of sitostanol (3 x Decreases in alkaline phosphatase and carotene concentration were 0.5 g/d). noted with sitosterol, but with sitostanol these returned to initial values. After 7 months of sitostanol, there was a slight but significant increase Safety was assessed by routine lab tests, and abdominal in transferrin concentration, and the total bile acid concentration in sonography of the liver and gallbladder were performed every 3 serum was significantly reduced. These changes did not reach months to check for possible gallstone formation. abnormal levels. All other lab values (hemoglobin, leukocyte count, platelet count, calcium, phosphate, lipase, iron, creatinine, and creatine kinase) showed no change. Weight and height percentiles of all children remained normal throughout the study. No changes in stool pattern were reported. No abnormalities in liver and gallbladder were found. 119 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Amundsen, Children consumed slightly less 2004287 spread (14 g/day vs 18 g/day during trial). Becker, 1993 All 9 subjects completed the 288 study. Children and parents reported a high degree of acceptance and compliance during all treatments. 120 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Clarke, 1990289 Increased incidence of epistaxis Descriptive study To evaluate the potentially beneficial effect of dietary 6 months in adolescents with familial Ontario, Canada fish oil supplementation on plasma lipid levels in a hypercholesterolemia treated with small group of adolescent patients with familial fish oil hyperlipoproteinemia type II, and to note possible undesirable side effects. Dennison, Randomized, double-blind randomized cross-over To assess whether psyllium was effect at lowering 4-5 weeks each 1993223 placebo-controlled, two-period trial of psyllium in children TC and LDL values in children beyond the reduction period with a 2 crossover clinical trial of psyllium ages 5-17 with LDL>110 achieved by diet alone week wash-out fiber in children with mg/dL already being period between hypercholesterolemia treated with diet, recruited from Pediatric Lipid Control Center and from private pediatricians in the vicinity. 121 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Clarke, 1990289 Diagnosis of FHL type II was based on a positive family history, typical 11 Age 11-21, abnormalities of plasma lipid levels, and absence of evidence of any other 63.6% female condition to account for the lipid abnormalities. All subjects were maintained on their usual modified low cholesterol/low sat. fat diet throughout the study; two also received colestipol, 10 g/day. Dennison, Children with LDL>110 mg/dL after treatment with diet. 25 children began the 11 males; 9 females 1993223 study; 4 did not return for age 11.1+3.8 years the final blood drawing (3 because of parental difficulties - timing, logistics; 1 boy because he refused to eat the cereal and have his blood redrawn); 1 second blood sample was lost. 122 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Clarke, 1990289 After 3 months of pretreatment observation, each subject was 3 of 11 subjects had prolongation of bleeding time during fish oil given a commercial fish oil supplement (containing 18% supplementation. 8 of the 11 subjects had a total of 9 episodes of eicosapentaenoic acid, 12% other omega-3 fatty acids, and epistaxis during 62 subject-months of fish oil treatment; no subject vitamin E) in a dosage started at 1 g/day, and increased by 1 reported any nosebleeds during 43 subject-months of pretreatment g/day at monthly intervals to a total of 5 g/day for the 5th and 6th and posttreatment observation (p<0.02). One subject experienced months of therapy. prolonged bleeding time ( 9 minutes; control, 3 to 5 minutes). The subject had been receiving 5 g/day of fish oil at the time of withdrawal. Two other subjects experienced epistaxis associated with modest prolongation of the bleeding time (7 minutes); one of these had a history of acetylsalicylic acid ingestion at the time. One subject had asymptomatic occult blood in the stool on one occasion; the bleeding time was normal. None of the subjects had any evidence of liver dysfunction; all prothrombin and partial thromboplastin times and platelet counts were normal. Dennison, Diet consisted of <30% calories as total fat, <10% as saturated One child (a 17 year old girl reported a transient increase in loose 1993223 fat and <200 mg dietary cholesterol/day. Total calories enough stools (3 /day) with the control cereal with resolved with discontinuation to ensure adequate growth. and reappeared with a second challenge of the control cereal and Subjects instructed to eat two 28gm servings (1 0z or 2/3 c each) resolved post discontinuation. of the control cereal (5 mg water-insoluble wheat fiber per serving) or the psyllium cereal (3 gm of water-insoluble fiber and 3 game water-soluble fiber per serving). Cereals were similar in No changes in subscapular or triceps skin-fold thicknesses. appearance and provided by same manufacturer. Height and weight increased equally during consumption of control and psyllium cereals. None of vitamin or mineral levels tested were lower after the psyllium cereal. 123 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Clarke, 1990289 In two subjects, therapy was stopped as a result of epistaxis. Dennison, Compliance was 82% for both baseline lipid levels not different 1993223 cereals. significantly for those (n=5) who did not complete the study. Most families were unaware of which cereal was in use (2/3 of children or parents incorrectly identified the psyllium containing cereal). 124 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Glassman, Treatment of Type Iia Descriptive study To determine whether the AHA Step-One diet would 8.1 (+- 2.4) months 1990290 hyperlipidemia in childhood by a Lipid clinic, New York effectively lower serum total and LDL cholesterol simplified American Heart levels in children for prolonged periods. Association (AHA) diet and fiber supplementation Gulesserian, Effect of a rapeseed oil Descriptive study To investigate how adolescents and families with FH 5 months 2002291 substituting diet on serum lipids Vienna, Austria would respond to a lipid lowering diet enriched with and lipoproteins in children and rapeseed oil and what effects on serum lipids can be adolescents with familial observed. hypercholesterolemia Gylling, 199539 Sitostanol ester margarine in Randomized cross-over To assess the effect of sitosterol rape-seed 6 weeks each arm dietary treatment of children with trial of diet margarine vs. rape-seed margarine without sitosterol familial hypercholesterolemia 14 children with HeFH Helsinki, Finland 125 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Glassman, Children with primary type Iia cholesterolemia (evaluated TC with normal or 36 Age range 9-17, mean 9.74 1990290 elevated TG concentrations) who were treated at the Children's Nutrition 38.9% female Center of the New York Medical College, Valhalla, between January 1 1988 and December31 1988. These patients were identified during routine screening by their pediatricians and had the diagnosis confirmed by lipoprotein electrophoresis after a 14-hour fast. The children were otherwise in excellent health and were receiving no medications. Gulesserian, HeFH patients were selected from a pediatric outpatient clinic with the 43 were enrolled, but Age range 4-19, median 2002291 following characteristics: plasma TC levels >200 mg/dL), LDL > 130 mg/dL, analysis includes only 17 12.7 TG <250 mg/dL, BMI range 10% to 90% percentiles. Of 43 potential subjects, 64.7% female 26 were excluded because they were not able to adhere to the study protocol or did not provide an appropriate dietary protocol. Gylling, 199539 Diagnosis of FH established in children and in one parent mostly by DNA 14 children with HeFH 7 boys, 7 girls technique and 1 child with HoFH (2 Mean age 9.1+1.1 (2-15) years old) 126 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Glassman, Patients were treated with a modified version of the AHA Step- All patients tolerated both regimens well without developing abdominal 1990290 One diet at the Children's Nutrition Center of the New York distention or cramping, constipation, diarrhea, or excessive flatus Medical College. The diet, in which a maximum of 10% of the production. There were no changes in serum zinc, copper, or energy intake was derived from saturated fat, did not specifically hemoglobin concentrations; hematocrit; mean corpuscular volume; or restrict cholesterol intake or the percentage of total energy prothrombin and partial thromboplastin times during the period of ingested as fat. In addition, children under age 7 were required study. to consume 2.5 g of soluble fiber as psyllium mixed in water twice daily. Children older than 7 received 10 g of psyllium. Each patient was reassessed at 6-week intervals using a 24-hour dietary diary, a food-selection checklist, and a serum lipoprotein electrophoresis. Serum zinc and copper levels, as well as prothrombin and partial thromboplastin times, were measured in 10 patients over the course of therapy. Gulesserian, Diet therapy had 2 approaches: 1) replacement of as many NR 2002291 visible fats as possible by rapeseed oil (including use preparation of meals, as in frying, baking, and salad dressing) and 2) reduction of dietary cholesterol. High dietary fiber was achieved by prescribing a high quantity of fruits and vegetables. During the study patients met the dietician 7 times, and the pediatrician 6 times. Compliance to the diet was checked by routine 3-day protocols with children and parents. Gylling, 199539 Children were randomized to replace 24 g of their normal daily Children noted no difference in taste between the two margarines. fat intake by the same amount of a rapeseed oil-rich margarine with or without sitostanol ester (M vs MS) . All children had Report "good compliance," and "well tolerated." previously been advised to use a low animal fat-low cholesterol diet rich in monogenic fatty acids for years. Protocol results in daily consumption of 3 gm of free sitostanol in MS group. 127 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Glassman, All patients tolerated both 1990290 regimens well. Gulesserian, The diet was well accepted; no 26 enrolled patients were not able to 2002291 patient canceled the study due adhere to the study protocol or did not to dislike of the oil. provide an appropriate dietary protocol. The report of compliance only includes the patients who adhered to the protocol. Gylling, 199539 Report "good compliance," and "well tolerated." 128 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Laurin, 1991294 Effects of a soy-protein beverage RCT, crossover study To examine the effects of a beverage based on 4 weektreatment/4 on plasma lipoproteins in children Quebec City isolated soy protein in the diet on plasma lipid, week washout/4 with familial hypercholesterolemia lipoprotein-fraction cholesterol, and apolipoprotein week crossover concentrations in HeFH children. treatment McCrindle, Garlic Extract Therapy in Children DB RCT: To determine tolerance, compliance, safety, and 8 weeks 1998235 with Hypercholesterolemia placebo-controlled trial of efficacy of therapy with a commercially available garlic extract garlic extract in lowering cholesterol in pediatric patients with hypercholesterolemia 30 patients 129 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Laurin, 1991294 Hypercholesterolemic children from the Quebec City area where the 10 Mean age 7.9 heterozygote frequency of FH is ~2.5-fold higher than in most populations 40% female elsewhere. Baseline fasting concentrations of TC and LDL > 95th percentile adjusted for age and sex. Other criteria: detection of hypercholesterolemia in at least 2 first-degree relatives; the presence of tendinous xanthomas or type II- A lipoprotein pattern in 3 generations of relatives, or with compatible autosomal dominant transmission; exclusion of secondary forms of hypercholesterolemia. No subjects had fasting hyperchylomicronemia or type III hyperlipoproteinemia, recent weight loss (within 6 months), acute disease or major surgery in the past 3 months, and allergy to or dislike for milk. McCrindle, Included patients age 8-18, a positive family history of hypercholesterolemia or 30 patients age 8-18 Mean age 14 1998235 premature atherosclerotic cardiovascular disease in first-degree relatives, a 53% male minimum fasting TC > 4.8 mmol/L (>185 mg/dL), participation in a dietary counseling program, and compliance with a NCEP Step II diet for at least 6 months. Excluded secondary causes of hyperlipidemia or a history of major surgery or serious illness within 3 months prior to enrollment. 130 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Laurin, 1991294 5 subjects per group were randomly assigned to receive either All participants had adequate growth and development as verified by the cow-milk or the soy-beverage diet, with subsequent switch to the tracking of each subject against physical-growth charts. the other diet in a crossover design, each diet and washout lasting 4 week. During the washout, subjects consumed their regular diet. The soy beverage was formulated to provide similar amounts of protein, carbohydrate, fat, calcium, and phosphorus as 2%-fat cow milk as well as a minimum of 20% of the US RDA of 16 other vitamins and minerals. The nutrient composition of the two experimental diets was comparable with their regular regimen, and the same menus were used for the two experimental periods. McCrindle, Patients were instructed to stop taking lipid-lowering medications 1 patient on garlic experienced unpleasant body odor, but had been 1998235 for >8 weeks prior to the study. Participants were instructed to working concomitantly on a garlic farm during the study period. take 1 capsule (containing 300 mg garlic extract or identical placebo) t.i.d. for 8 weeks. A random number generated list was The most commonly reported AEs were headache and upset stomach, supplied to an independent pharmacist who assigned patients but reporting of AEs was similar between garlic and placebo groups. consecutively when they were enrolled. Logbooks and questionnaires were used. There were no significant differences between groups in height, weight, blood pressure. The only significant differences in laboratory parameters were in serum albumin level (+2.0 g/L, p=0.002) and hemoglobin (+5.2 g/L, p=0.02) with garlic. 1 patient was unable to swallow the capsules and withdrew. Compliance was similar between garlic and placebo groups. 131 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Laurin, 1991294 Authors report that good control over mean dietary intake was achieved, as indicated by similar percentages of protein energy consumed as a dairy source in the two experimental diets. McCrindle, 1 patient was unable to swallow 1998235 the capsules and withdrew. Compliance was similar between garlic and placebo groups. 86% vs 93%, (p=0.34). 132 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Mietus-Snyder, Endothelial dysfunction occurs in Non-randomized To confirm the previously-reported finding of 6 weeks 1998292 children with two genetic controlled trial impaired endothelium-dependent vascular relaxation hyperlipidemias: Improvement 33 of 45 subjects were in children with FH, to extend these observations to with antioxidant vitamin therapy members of families other hyperlipidemic phenotypes and to assess the followed in a lipid clinic. efficacy of antioxidant vitamin therapy in reducing Control Group: endothelial dysfunction. Unaffected siblings or children with normal lipid profiles Sanchez-Bayle, The effect of fiber Observational study, Spain To assess the effects of fiber supplements on lipid 3 months 2001293 supplementation on lipid profile in levels in children with dyslipidemia who were children with following a dietary treatment. hypercholesterolemia Schlierf, Sitosterol in juvenile type II Randomized cross-over To determine the effect of verum to placebo 6 months, 3 on 1978295 hyperlipoproteinemia trial comparing verum sitosterol (sitosterol) to placebo; 3 months each treatment. West Germany 133 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Mietus-Snyder, 1. Good health and diagnosed with heterozygous FH (LDL > 95th percentile, 45 Mean age: 12.5 1998292 normal triglycerides, affected parent) or familial combined 11 received treatment 56% female hyperlipoproteinemia (LDL or fasting triglycerides > 95th percentile, parent with No withdrawals reported 1 of 3 phenotypic presentations of FCH). 18 subjects with FH 15 subjects with FCH 12 controls Sanchez-Bayle, Children with persistently elevated LDL (>135 mg/dL) after a minimum of 6 53 children in Madrid, Mean age 7.3 (range 4-18) 2001293 months dietary treatment (<30% calories as total fat, <10% as sat. fat, and Spain <300 mg dietary cholesterol per day) were recruited from Hospital Nino Jesus in Madrid from 1989 to 1990. As a control group, 33 children with persistently elevated LDL were recruited from 1992 to 1993. Schlierf, NR 15 children and Age 8-20 1978295 adolescents, randomized to receive either sitosterol (n=7)or placebo first (n=8). 134 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Mietus-Snyder, Exposures: Combined regimen of partial difference-alpha No adverse effects of antioxidant vitamin therapy reported. 1998292 tocopherol (400 IU twice a day) and vitamin C (500 mg twice a day) for 6 weeks. At follow-up the protocol was repeated except for nitroglycerin administration. Brief medical history, physical exam, and nutritional counseling including American Heart Association diet guidelines. Vascular reactivity in brachial artery measured with high- resolution two-dimensional ultrasonography. Primary outcome was change in brachial artery diameter in response to either hyperemia or sublingual administration of nitroglycerin Sanchez-Bayle, In addition to dietary restriction, children were given fiber Two children reported abdominal discomfort and soft stools and 2001293 supplements for 3 months in tablets that contained 50% wheat refused to continue the study. bran and 50% pectin and were given 2 to 3 times per day at doses of 50 mg/kg/day. Compliance with diet and fiber supplementation was confirmed by interview. Diet was assessed by a food frequency questionnaire and a 24-hour recall. Control families were instructed to maintain their current dietary restriction and subjects were re-evaluated 3 months later. Schlierf, Treatment: 12g of B-sitosterol granulate with 20g NaBr 3 patients terminated the study before completion, unclear reasons; in 1978295 Placebo: 20g of NaBr with 12 g cellulose. Each taken in 3 the other 12 plasma bromide levels, drug intake sheets and used med portions of 4g, 30 minutes before meals. packs indicated good compliance. Sitosterol levels 0.88+0.24 in placebo group vs 1.48+0.62 in sitosterol group. 135 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Mietus-Snyder, "All subjects tolerated the study Small sample size 1998292 well, although approximately 1/3 Short follow-up complained of a transient, mild headache after nitroglycerin administration, coincident with a 10 to 20 mm Hg decrease in systolic pressure." Sanchez-Bayle, Fiber supplements were 2001293 generally well tolerated except for two children who withdrew due to abdominal discomfort. Schlierf, 3 of the 15 patients withdrew Poor quality 1978295 before completion. In the results lumped after 6 months (no pre- remaining 12 patients, plasma cross over data). bromide levels, drug intake No wash-out period. control sheets, and the returned Authors comment that this not probably packs of unused medication recommended for use in children indicated good adherence to the because of increases in HDL and study protocol. sitosterol levels. 136 Evidence Table 5. Adverse Effects of Interventions Author, year Title Type of Study/Setting Aims Duration of Study Zavoral, The hypolipidemic effect of locust Non-randomized, To assess the hypolipidemic effect of locust bean 8 weeks 1983296 bean gum food products in crossover study gum food products. familial hypercholesterolemic adults and children Exercise Kavey, 1997298 Valvular and pediatric congenital Cross-sectional study, US To compare the blood pressure response to exercise 0 (cross-sectional heart disease - exaggerated blood in a group of children with increased LDL with the study) pressure response to exercise in response in a group of normal control subjects. children with increased low- density lipoprotein cholesterol Tolfrey, 1998297 Exercise training induced Observational pre- Examine effect of exercise training on prepubertal 12 weeks alterations in prepubertal test/post-test children's lipid-lipoprotein profile children's lipid-lipoprotein profile Recruitment from primary schools England February - July 1995 (ET) February - July 1996 (Control) 137 Evidence Table 5. Adverse Effects of Interventions Author, year Main Eligibility Criteria Enrolled Demographics Zavoral, Normal and familial hypercholesterolemia patients were selected from the 11 children Aged 10-18 1983296 Hennepin County Medical Center's Hyperlipidemia Clinic. All of the families had one parent and child and at least one additional first degree relative with FHC. Exercise Kavey, 1997298 The study group consisted of boys aged >10 with LDL >160 mg/dL who had 15 hypercholesterolemic Age 10-17, mean 12.7 been referred to the Pediatric Preventive Cardiology clinic over a 2-year children; 32 control 0% female period. Higher SBP on termination of exercise in these children was children evaluated by using a retrospective control group consisting of boys 10-18 with normal results on cardiac examination who had undergone elective treadmill exercise testing during the same period. Children with relative weight >120% were excluded from both groups. A prospective evaluation was also undertaken: 10 hypercholesterolemic boys with LDL >160 mg/dL and 10 age-matched control subjects age 10-17 were examined by the same protocol. Tolfrey, 1998297 1. Volunteered for study after attending school-based presentation and visiting 48 enrolled Mean age: 10.6 (0.7) university exercise physiology lab. 2. Determined by medical screening to be ET group: 28 50% female healthy and asymptomatic. 3. No medication known to effect lipid profile. Control group: 20 138 Evidence Table 5. Adverse Effects of Interventions Author, year Interventions/Exposures Adverse Effects Zavoral, Identical food products with and without locust bean gum (LBG) There was no change observed in the white blood cell counts, serum 1983296 were consumed by two groups of arbitrarily assigned patients calcium, or SGOT determinations. using a cross-over design. The food products contained 8 to Subjects reported increased rectal gas while on the LBG-containing 10% LBG by dry weight. Each portion contained between 1.7 products, but decreased and at 1-2 weeks the amount of gas reported and 2.0 g of LBG. LBG in the food products was prescribe to the was usually not a problem. No one had diarrhea, and stool size subjects on the basis of body size. Families were seen at 2- increased slightly. One patient had significant constipation that week intervals, and dieticians collected 3-day diet histories and remitted after discontiuing the diet for 3 days. This subject was able to 2week dietary records for the food products. Safety resume the diet without further problems. assessments included WBC counts, calcium, and SGOT before and after feedbing LBG food products. Exercise Kavey, 1997298 Treadmill exercise tests were performed on a cardioexercise Combined results for retrospective and prospective comparisons: treadmill, using published protocol and pediatric normal data for Children with severely increased LDL cholesterol had an exaggerated maximum predicted heart rate and endurance time. Blood blood pressure response to exercise when compared with pressures were measured in the right arm at rest before normolipidemic control subjects. Children with increased LDL had beginning exercise, immediately after termination of exercise, significantly higher systolic and diastolic blood pressures before and then every 2 minutes during recovery. treadmill exercise (S 120 vs 113 mmHG, p<0.03; D 68 vs 63 mmHg, p<0.01). Among children with high LDL, blood pressures were significantly higher after exercise (S 182 vs 160 mmHg, p<0.0003; D 77 vs 72 mmHg, p<0.03), and systolic blood pressures remained significantly higher at the end of recovery (120 vs 112 mmHg, p<0.005. Tolfrey, 1998297 Exercise Training Group (ET): NR Supervised stationary cycling for 30 minutes, 3x/week for 12 weeks at 79.3% (+ 1.2) peak heart rate Control group: Maintained their usual lifestyle pattern 139 Evidence Table 5. Adverse Effects of Interventions Author, year Compliance/Tolerance Comment Zavoral, Children ate the prescribed 1983296 amount of LBG food products. Author states "LBG food acceptance was good, and there were no significant side effects." Exercise Kavey, 1997298 NR Tolfrey, 1998297 100% adherence to training program reported 140 Table 18. Summary of Systematic Evidence Review Level and Type Quality of Arrow Key question of Evidence Evidence Conclusions 9 What are the adverse effects I, II-1, II-2 Fair Controlled and non-controlled studies of treatment reported of drug, diet, exercise, and adverse effects of drug, diet, exercise, and combination combination therapy in therapy in children and adolescents. Statin drugs were children/adolescents? associated primarily with elevations in LFTs and CK. Bile-acid binding resins were associated with GI side effects and decreased levels of serum vitamins and minerals. Low fat diet has been associated with growth retardation and nurtional dwarfing in 3 children who were placed on low-fat diets without formal advice and monitoring. Most studies show normal growth and development in children over 2 years old on monitored low-fat diets. Few side effects other than elevated blood pressure were noted with exercise. The duration of follow-up in these studies ranged from 10 days to 8 years. Studies were generally not of sufficient duration to determine long-term effects of either short or extended use. 10 Does improving dyslipidemia None in childhood reduce the risk of dyslipidemia in adulthood? 11 What are the cost issues None involved in screening for dyslipidemia in asymptomatic children? Abbreviations CHD=Coronary heart diesease, CK=Creatine kinase, FH=Familial hyperlipidemia, FCH=Familial Combined Hyperlipidemia, HDL=High-density Lipoprotein, LDL=Low-density Lipoprotein, LFT=Liver function test, NCEP=National Cholesterol Education Program, RCT=Randomized controlled trial, TC=Total cholesterol, TG=Triglycerides 4 Appendix 1. Abbreviations AAP: American Academy of Pediatrics AE: Adverse Event BL: Baseline BMI: Body Mass Index [kg/m2] CAD: Coronary Artery Disease CHD: Coronary Heart Disease CVD: Cardiovascular Disease DB: Double Blind EDRF/NO: Endothelium-derived Relaxing Factor/Nitric Oxide ET: Treatment Group FCH: Familial Combined Hyperlipidemia FCHL: Familial Combined Hyperlipoproteinemia FH: Familial Hypercholesterolemia FHTG: Familial Triglyceridemia FMD: Flow Mediated Dilation HC: Hypercholesterolemia HDL: High Density Lipoprotein HDL-C: High Density Lipoprotein Cholesterol HeFH: Heterozygous Familial Hypercholesterolemia HoF: Homozygous Familial Hypercholesterolemia HR: Heart Rate HTN: Hypertension IMT: Intimal-medial Thickness LDL: Low-density Lipoprotein LDL-C: Low-density Lipoprotein Cholesterol LRC: Lipid Research Clinic MI: Myocardial Infarction NA: Not Applicable NCEP: National Cholesterol Education Program NIH: National Institute of Health NR: Not Reported NS: Not Significant OCP: Oral Contraceptive Pills RCT: Randomized Controlled Trial RF: Risk Factor RR: Risk Ratio SEM: Standard Error of Measure SSF: Sum of Skinfolds TC: Total Cholesterol TG: Triglycerides VLDL: Very Low-density Lipoprotein 1 Appendix 2. Units of Measure Conversion Formulas Conversion for TC (total cholesterol), HDL (high density lipoprotein) and LDL (low density lipoprotein)* †: To get from SI units (mmol/L) to mg/dL multiply by 38.67. To get from mg/dL to SI (in mmol/L) multiply by 0.02586. Conversion for TG (triglycerides)* †: To get from SI units (mmol/L) to mg/dL multiply by 88.57. To get from mg/dL to SI units, multiply by 0.01129. *References: www.ulb.ac.be/eraseme/edu/gastrocd/convers.htm www.fatfreekitchen.com/cholesterol/cholesterol_units.html www.unc.edu/~rowlett/units/scales/clinical_data.html www.globalrph.com/conv_si.htm www.nephron.com/cgi-bin/SI.cgi (conversion calculator) These accessed December 15, 2004 †Note: for both TC and TG conversions, some people use rounded conversion factors to get from mg/dL to SI units (i.e. 0.0259 and 0.0113). 1 Appendix 3. Search Strategies for Child and Adolescent Dyslipidemia Adverse Effects Database: Ovid MEDLINE(R) <1966 to September 2005> Search Strategy: ------1 exp Hyperlipidemia/ 2 ((adverse$ adj effect$) or harm$ or stigma$ or prejudice$ or discriminate$). mp. or exp stress, psychological/et or exp life change events/ or exp prejudice/ or exp stereotyping/ or exp self concept/ 3 exp "Wounds and Injuries"/et [Etiology] 4 exp hyperlipidemia/dt 5 exp Antilipemic Agents/ae, ct, to 6 1 and (2 or 3) 7 4 and 5 8 6 or 7 9 limit 8 to English language 10 8 not 9 11 limit 10 to abstracts 12 9 or 11 13 limit 12 to "all child (0 to 18 years)" Cost Database: Ovid MEDLINE(R) <1966 to September 2005> Search Strategy: ------1 exp HYPERLIPIDEMIA/ 2 exp lipoproteins/bl 3 1 or 2 4 Mass Screening/ 5 exp hyperlipidemia/di 6 3 and 4 7 5 or 6 8 exp "Costs and Cost Analysis"/ 9 7 and 8 10 exp HYPERLIPIDEMIA/ec [Economics] 11 9 or 10 12 limit 11 to all children <0 to 18 years> 13 limit 12 to English language Diet Database: Ovid MEDLINE(R) <1966 to September 2005> Search Strategy: ------ 1 Appendix 3. Search Strategies for Child and Adolescent Dyslipidemia 1 exp HYPERLIPIDEMIA/dh [Diet Therapy] 2 exp HYPERLIPIDEMIA/ 3 exp Diet/ 4 exp Diet Therapy/ 5 3 or 4 6 2 and 5 7 1 or 6 8 limit 7 to human 9 limit 8 to all children <0 to 18 years> 10 limit 9 to English language 11 9 not 10 12 limit 11 to abstracts 13 10 or 12 Drug Database: Ovid MEDLINE(R) <1966 to September 2005> Search Strategy: ------1 exp HYPERLIPIDEMIA/dt [Drug Therapy] 2 exp Antilipemic Agents/ad, ae, ct, tu, to [Administration & Dosage, Adverse Effects, Contraindications, Therapeutic Use, Toxicity] 3 exp HYPERLIPIDEMIA/ 4 2 and 3 5 1 or 4 6 limit 5 to all children <0 to 18 years> 7 limit 6 to English language 8 6 not 7 9 limit 8 to abstracts 10 7 or 9 Exercise Database: Ovid MEDLINE(R) <1966 to September 2005> Search Strategy: ------1 exp HYPERLIPIDEMIA/ 2 exp lipoproteins/bl 3 1 or 2 4 exp Exercise Movement Techniques/ 5 exercis$.mp. 6 (physical$ adj3 active$).mp. [mp=title, original title, abstract, name of substance, mesh subject heading] 7 (physical$ adj (fit or fitness)).mp. [mp=title, original title, abstract, name of substance, mesh subject heading] 8 4 or 5 or 6 or 7 2 Appendix 3. Search Strategies for Child and Adolescent Dyslipidemia 9 3 and 8 10 limit 9 to all child <0 to 18 years> 11 limit 10 to English language 12 10 not 11 13 limit 12 to abstracts 14 11 or 13 Family History Database: Ovid MEDLINE(R) <1966 to September 2005> Search Strategy: ------1 exp HYPERLIPIDEMIA/ge [Genetics] 2 exp HYPERLIPIDEMIA/ 3 exp lipoproteins/bl 4 2 or 3 5 exp Disease Susceptibility/ 6 4 and 5 7 1 or 6 8 exp RISK/ 9 7 and 8 10 limit 9 to all child <0 to 18 years> 11 (family$ adj3 histor$).mp. [mp=title, original title, abstract, name of substance, mesh subject heading] 12 4 and 11 13 limit 12 to all child <0 to 18 years> 14 13 and 8 15 10 or 14 16 limit 15 to English language 17 15 not 16 18 limit 17 to abstracts 19 16 or 18 Outcomes Database: Ovid MEDLINE(R) <1966 to September 2005> Search Strategy: ------1 exp Hyperlipidemia/co, dh, dt, th 2 exp Cerebrovascular Disorders/mo, pc, ep, et [Mortality, Prevention & Control, Epidemiology, Etiology] 3 exp Myocardial Ischemia/mo, pc, ep, et [Mortality, Prevention & Control, Epidemiology, Etiology] 4 1 and 2 5 1 and 3 6 4 or 5 3 Appendix 3. Search Strategies for Child and Adolescent Dyslipidemia 7 exp "OUTCOME AND PROCESS ASSESSMENT (HEALTH CARE)"/ 8 exp prognosis/ 9 7 or 8 10 1 and 9 11 6 or 10 12 limit 11 to all child <0 to 18 years> 13 limit 12 to English language 14 12 not 13 15 limit 14 to abstracts 16 13 or 15 Screening Database: Ovid MEDLINE(R) <1966 to September 2005> Search Strategy: ------1 exp HYPERLIPIDEMIA/ 2 exp Lipoproteins/bl [Blood] 3 Cholesterol/bl [Blood] 4 Triglycerides/bl [Blood] 5 2 or 3 or 4 6 exp Mass Screening/ 7 5 and 6 8 1 and 6 9 7 or 8 10 limit 9 to all child <0 to 18 years> 11 limit 10 to English language 12 10 not 11 13 limit 12 to abstracts 14 11 or 13 Sensitivity Database: Ovid MEDLINE(R) <1966 to September 2005> Search Strategy: ------1 exp HYPERLIPIDEMIA/ 2 exp Lipoproteins/bl [Blood] 3 Cholesterol/bl [Blood] 4 Triglycerides/bl [Blood] 5 2 or 3 or 4 6 1 or 5 7 exp "Sensitivity and Specificity"/ 8 exp Diagnostic Errors/ 9 7 or 8 10 6 and 9 4 Appendix 3. Search Strategies for Child and Adolescent Dyslipidemia 11 limit 10 to all child <0 to 18 years> 12 limit 11 to English language 13 11 not 12 14 limit 13 to abstracts 15 12 or 14 5 Appendix 4. Inclusion/Exclusion Criteria and Results of Searches Number of Number of Articles Abstracts Reviewed Number of Screened from Articles with Abstracts Included in Key Applicable Abstract Level Inclusion / Eligibility and other Evidence Questions Searches Exclusion Criteria Paper Level Inclusion / Exclusion Criteria Criteria*† Sources*†‡ Synthesis*† Screening 2 screening Include: Include: 780 369 160 family history RCTs, observational studies, cross- RCTs, observational studies, cross-sectional sensitivity sectional studies, practice studies, practice guidelines, meta analyses, or guidelines, meta analyses, or other other reviews; reviews; Examines screening for dyslipidemia in Examines screening for children and adolescents, specifically family dyslipidemia in children or history, biological (obesity, insulin level, etc) adolescents, specifically risk and lifestyle (physical activity, diet, etc) risk factors, adverse effects of factors, measurement of serum lipids (TC, screening, tests or test HDL, LDL, TG), tests or test performance performance; (sensitivity, specificity), other screening Information on background or cost measurements, adverse effects of screening; Information on background or cost; Exclude: For prevalence, normal values papers: Wrong publication type, or include non-U.S.-based if n>100 population; For neonatal screening and testing papers: Not relevant to topic, non- include only if has mutation analysis U.S./Western culture, or non- English language; Exclude: Examines metabolic syndrome, Wrong publication type, or population; triglyceride specific outcomes, Not relevant to topic, non-U.S./Western screening in secondary dyslipidemia culture, or non-English language; Examines metabolic syndrome, triglyceride specific outcomes, screening in secondary dyslipidemia 1 Appendix 4. Inclusion/Exclusion Criteria and Results of Searches Number of Number of Articles Abstracts Reviewed Number of Screened from Articles with Abstracts Included in Key Applicable Abstract Level Inclusion / Eligibility and other Evidence Questions Searches Exclusion Criteria Paper Level Inclusion / Exclusion Criteria Criteria*† Sources*†‡ Synthesis*† Interventions 4-8, 10 diet Include: Include: 1668 707 68 exercise Treatment trials, longitudinal RCTS, clinical controlled trials, non-controlled drug studies, natural history studies, trials, observational studies, cross sectional outcomes cross sectional studies, practice studies, practice guidelines, meta analyses, or guidelines, meta analyses, or other other reviews; reviews; Examines interventions for dyslipidemia in Examines interventions for children and adolescents, specifically drugs, dyslipidemia in children or diets, exercise, or combinations, lipid adolescents, specifically drugs, outcomes, other outcomes (intimal thickness, diets, exercise or combinations, medial dilation, etc), adverse effects of family history, risk factors, adverse interventions, tests and test performance; effects of interventions, tests or test Information on background or cost; performance; For natural history papers: include non- Information on background or cost; Western culture For natural history papers: include non-Western culture Exclude: Wrong population type, or population (< 50% Exclude: child/adolescent or no separate analysis for Wrong publication type or child/adolescent); population; Not relevant to topic, non-U.S./Western Not relevant to topic, non- culture (except natural history papers), or non- U.S./Western culture (except English language; natural history papers), or non- Examines metabolic syndrome, triglyceride English language; specific outcomes, screening in secondary Examines metabolic syndrome, dyslipidemia triglyceride specific outcomes, screening in secondary dyslipidemia 2 Appendix 4. Inclusion/Exclusion Criteria and Results of Searches Number of Number of Articles Abstracts Reviewed Number of Screened from Articles with Abstracts Included in Key Applicable Abstract Level Inclusion / Eligibility and other Evidence Questions Searches Exclusion Criteria Paper Level Inclusion / Exclusion Criteria Criteria*† Sources*†‡ Synthesis*† Adverse Effects of Screening 3 adverse same as KQ 2 same as KQ 2 163 84 8 effects Adverse Effects of Interventions 9 adverse same as KQ 1, 4-8, 10 Include: same as same as 81 effects Treatment trials, longitudinal studies, natural KQ 3 KQ 3 history studies, meta analyses, or other reviews Examines adverse effects of dyslipidemia in children and adolescents, specifically from drugs, diets, exercise, or combinations; Information on background or cost Exclude: Wrong publication type, or population; Not relevant to topic, non-U.S./Western culture, or non-English language 3 Appendix 4. Inclusion/Exclusion Criteria and Results of Searches Number of Number of Articles Abstracts Reviewed Number of Screened from Articles with Abstracts Included in Key Applicable Abstract Level Inclusion / Eligibility and other Evidence Questions Searches Exclusion Criteria Paper Level Inclusion / Exclusion Criteria Criteria*† Sources*†‡ Synthesis*† Cost 11 cost Include: same as cost abstract criteria 26 15 7 Study of cost of screening and/or treatment for elevated cholesterol in children; Information on background Exclude: Genetic screening or tracking study based on identified FH individual; Wrong population; Not relevant to topic, non- U.S./Western culture , or non- English language *All abstracts were reviewed for applicability to other key questions. †Duplicates may exist between key questions. ‡Other sources include reference lists, experts, etc. 4 Appendix 5. U.S. Preventive Services Task Force Quality Rating Criteria* Diagnostic Accuracy Studies Criteria • Screening test relevant, available for primary care, adequately described • Study uses a credible reference standard, performed regardless of test results • Reference standard interpreted independently of screening test • Handles indeterminate results in a reasonable manner • Spectrum of patients included in study • Sample size • Administration of reliable screening test Definition of ratings based on above criteria Good: Evaluates relevant available screening test; uses a credible reference standard; interprets reference standard independently of screening test; reliability of test assessed; has few or handles indeterminate results in a reasonable manner; includes large number (more than 100) broad-spectrum patients with and without disease. Fair: Evaluates relevant available screening test; uses reasonable although not best standard; interprets reference standard independent of screening test; moderate sample size (50 to 100 subjects) and a “medium” spectrum of patients. Poor: Has important limitation such as: uses inappropriate reference standard; screening test improperly administered; biased ascertainment of reference standard; very small sample size of very narrow selected spectrum of patients. Randomized Controlled Trials (RCTs) and Cohort Studies Criteria • Initial assembly of comparable groups: RCTs—adequate randomization, including concealment and whether potential confounders were distributed equally among groups; cohort studies—consideration of potential confounders with either restriction or measurement for adjustment in the analysis; consideration of inception cohorts • Maintenance of comparable groups (includes attrition, cross-overs, adherence, contamination) • Important differential loss to follow-up or overall high loss to follow-up • Measurements: equal, reliable, and valid (includes masking of outcome assessment) • Clear definition of interventions • Important outcomes considered • Analysis: adjustment for potential confounders for cohort studies, or intension-to- treat analysis for RCTs 1 Appendix 5. U.S. Preventive Services Task Force Quality Rating Criteria Definition of ratings based on above criteria Good: Meets all criteria: Comparable groups are assembled initially and maintained throughout the study (follow-up at least 80 percent); reliable and valid measurement instruments are used and applied equally to the groups; interventions are spelled out clearly; important outcomes are considered; and appropriate attention to confounders in analysis. Fair: Studies will be graded “fair” if any or all of the following problems occur, without the important limitations noted in the “poor” category below: Generally comparable groups are assembled initially but some question remains whether some (although not major) differences occurred in follow-up; measurement instruments are acceptable (although not the best) and generally applied equally; some but not all important outcomes are considered; and some but not all potential confounders are accounted for. Poor: Studies will be graded “poor” if any of the following major limitations exists: Groups assembled initially are not close to being comparable or maintained throughout the study; unreliable or invalid measurement instruments are used or not applied at all equally among groups (including not masking outcome assessment); and key confounders are given little or no attention. Case Control Studies Criteria • Accurate ascertainment of cases • Nonbiased selection of cases/controls with exclusion criteria applied equally to both • Response rate • Diagnostic testing procedures applied equally to each group • Measurement of exposure accurate and applied equally to each group • Appropriate attention to potential confounding variable Definition of ratings based on criteria above Good: Appropriate ascertainment of cases and nonbiased selection of case and control participants; exclusion criteria applied equally to cases and controls; response rate equal to or greater than 80 percent; diagnostic procedures and measurements accurate and applied equally to cases and controls; and appropriate attention to confounding variables. Fair: Recent, relevant, without major apparent selection or diagnostic work-up bias but with response rate less than 80 percent or attention to some but not all important confounding variables. 2 Appendix 5. U.S. Preventive Services Task Force Quality Rating Criteria Poor: Major selection or diagnostic work-up biases, response rates less than 50 percent, or inattention to confounding variables. *Reference: Harris RP, Helfand M, Woolf SH, Lohr KN, Mulrow CD, Teutsch SM et al. Current methods of the U.S. Preventive Services Task Force: a review of the process. Am J Prev Med 2001; 20(3 Suppl):21-35. 3 Appendix 6. Statistical Methods Used for Meta Analysis Steps to get mean percentage change: 1. Use reported mean change. 2. If not reported, calculated using mean lipid level changes from endpoint to baseline of the treatment and control group. Steps to get standard error (SE) for mean percentage change: 1. Use reported SE. 2. If SE is not reported and standard deviation (SD) is reported, calculated from SD. 3. If both SE and SD not reported, calculated from 95% CI if reported. 4. If none of above is reported, calculated from the reported mean lipid level change and the associated variance (or SD). Let x and y denote the mean lipid level change from endpoint to baseline of the control and treatment group, respectively, the variance of mean percentage change is given by 2 y − x y⎛⎞ y⎛⎞ Var() x Var () y Var()== Var ()⎜⎟⎜⎟22 + xxx⎝⎠⎝⎠xy using delta method. yx− Let z = , where x and y denote the individual lipid level change from x endpoint to baseline, then mean percentage change is given by the mean zz, . It y could be shown that Var() could be approximated by Var() z using delta- x method. 5. If SD for endpoint is not available, use estimates of SD from baseline (or maximum estimates for HDL) to get a conservative estimate of SE for mean percent change. 1